Hi, Jeff!
Recently i got 2.4.0 crashed with oops in rtl8139_rx_interrupt() due
to packet with size = 0.
Patch and oops message below.
Cheers,
-up
patch-8139driver-2.4.0
--- linux-2.4.0/drivers/net/8139too.c Mon Jan 22 18:14:22 2001
+++ linux/drivers/net/8139too.c Mon Jan 22 18:20:14 2001
@@ -210,6 +210,9 @@ static int multicast_filter_limit = 32;
/* max supported ethernet frame size -- must be at least (dev->mtu+14+4).*/
#define MAX_ETH_FRAME_SIZE 1536
+/* min supported ethernet frame size, according to RFC894 */
+#define MIN_ETH_FRAME_SIZE 60
+
/* Size of the Tx bounce buffers -- must be at least (dev->mtu+14+4). */
#define TX_BUF_SIZE MAX_ETH_FRAME_SIZE
#define TX_BUF_TOT_LEN (TX_BUF_SIZE * NUM_TX_DESC)
@@ -1804,7 +1807,8 @@ static void rtl8139_rx_interrupt (struct
* Rx processing.
*/
if ((rx_size > (MAX_ETH_FRAME_SIZE+4)) ||
- (!(rx_status & RxStatusOK))) {
+ (!(rx_status & RxStatusOK)) ||
+ (rx_size < (MIN_ETH_FRAME_SIZE + 4))) {
rtl8139_rx_err (rx_status, dev, tp, ioaddr);
return;
}
OOPS MESSAGE:
Unable to handle kernel paging request at virtual address 0xcc000000
*pde = 00000000
Oops: 00000000
CPU: 0
EIP: 0010:[<ce89f1a4>]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010206
eax: cb798aa0 ebx: cb798aa0 ecx: 3fe94dc4 edx: fffffffc
esi: cc000000 edi: cb86b77e ebp: fffffffc esp: cab45d1c
ds: 0018 es: 0018 ss: 0018
Stack: 00000051 ce8a209e ce8a2000 cbd57c00 ce8a2038 ce8a2037 00000000 ce893710
cba50000 ce89f63b cbd57c00 cbd57d40 ce8a2000 cbd4faa0 04000001 0000000b
cab45db4 00000000 00000014 cbd57d40 c010a05d 0000000b cbd57c00
cab45db4
Call Trace: [<ce8a203e>] [<ce8a2000>] [<ce8a2038>] [<ce8a2037>] [<ce893710>]
[<ce8a2000>] [<c010a05d>] [<c010a1ce>] [<c0108e6c>] [<c01189ba>] [<c010a201>]
[<c0108e6c>] [<c0180018>] [<c018a290>] [<c018a81b>] [<c017d686>] [<c017d4f1>]
[<c017f2a4>] [<c017b384>] [<c017f168>] [<c012fd32>] [<c0108dc3>]
Code: f3 a5 f6 c2 02 74 02 66 a5 f6 c2 01 74 01 a4 89 e8 03 83 80
>>EIP; ce89f1a4 <[8139too]rtl8139_rx_interrupt+164/264> <=====
Trace; ce8a203e <.data.end+163f/????>
Trace; ce8a2000 <.data.end+1601/????>
Trace; ce8a2038 <.data.end+1639/????>
Trace; ce8a2037 <.data.end+1638/????>
Trace; ce893710 <[opl3sa2]__module_author+9570/eec0>
Trace; ce8a2000 <.data.end+1601/????>
Trace; c010a05d <handle_IRQ_event+31/5c>
Trace; c010a1ce <do_IRQ+6e/b4>
Trace; c0108e6c <ret_from_intr+0/20>
Trace; c01189ba <do_softirq+3e/6c>
Trace; c010a201 <do_IRQ+a1/b4>
Trace; c0108e6c <ret_from_intr+0/20>
Trace; c0180018 <mmap_mem+40/c4>
Trace; c018a290 <vt_console_print+134/2e4>
Trace; c018a81b <vc_init+3f/11c>
Trace; c017d686 <opost_block+186/194>
Trace; c017d4f1 <opost+1a5/1b4>
Trace; c017f2a4 <write_chan+13c/1e4>
Trace; c017b384 <tty_write+1b8/1fc>
Trace; c017f168 <write_chan+0/1e4>
Trace; c012fd32 <sys_write+8e/c4>
Trace; c0108dc3 <system_call+33/38>
Code; ce89f1a4 <[8139too]rtl8139_rx_interrupt+164/264>
00000000 <_EIP>:
Code; ce89f1a4 <[8139too]rtl8139_rx_interrupt+164/264> <=====
0: f3 a5 repz movsl %ds:(%esi),%es:(%edi) <=====
Code; ce89f1a6 <[8139too]rtl8139_rx_interrupt+166/264>
2: f6 c2 02 testb $0x2,%dl
Code; ce89f1a9 <[8139too]rtl8139_rx_interrupt+169/264>
5: 74 02 je 9 <_EIP+0x9> ce89f1ad
<[8139too]rtl8139_rx_interrupt+16d/264>
Code; ce89f1ab <[8139too]rtl8139_rx_interrupt+16b/264>
7: 66 a5 movsw %ds:(%esi),%es:(%edi)
Code; ce89f1ad <[8139too]rtl8139_rx_interrupt+16d/264>
9: f6 c2 01 testb $0x1,%dl
Code; ce89f1b0 <[8139too]rtl8139_rx_interrupt+170/264>
c: 74 01 je f <_EIP+0xf> ce89f1b3
<[8139too]rtl8139_rx_interrupt+173/264>
Code; ce89f1b2 <[8139too]rtl8139_rx_interrupt+172/264>
e: a4 movsb %ds:(%esi),%es:(%edi)
Code; ce89f1b3 <[8139too]rtl8139_rx_interrupt+173/264>
f: 89 e8 movl %ebp,%eax
Code; ce89f1b5 <[8139too]rtl8139_rx_interrupt+175/264>
11: 03 83 80 00 00 00 addl 0x80(%ebx),%eax
Kernel panic: Aiee, killing interrupt handler
580 warnings issued. Results may not be reliable.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
