Silent data corruption on em(4) interfaces

[KOT MATPOCKuH]

Hello!

I got silent data corruption when transferring data via em(4) interface on
10.2-STABLE r286912.
1. I got broken large file transferred via ftp (MD5 checksum mismatched);
2. I got disconnects when transferring large data via ssh with messages:
Corrupted MAC on input.
Disconnecting: Packet corrupt

Problem occurs only after few hours of uptime. Immediately after reboot I
transferred same file via ftp without any errors.

I tried to use:
- em0 and em2 interfaces in link aggregation
- em1 as "clean" interface
But I got same problem in both cases.

Also one time when transferring file I got this messages:
em0: Interface stopped DISTRIBUTING, possible flapping
em0: Watchdog timeout -- resetting
em2: Interface stopped DISTRIBUTING, possible flapping
em2: Watchdog timeout -- resetting

netstat -in does not see any problems:
Name    Mtu Network       Address              Ipkts Ierrs Idrop    Opkts
Oerrs  Coll
em0    1500 <Link#1>      00:14:4f:01:3f:7a  6689452     0     0
146720     0     0
em1    1500 <Link#2>      00:14:4f:01:3f:7b  5732168     0     0
2865912     0     0
em2    1500 <Link#3>      00:14:4f:01:3f:7c   501817     0     0
3392333     0     0

Network adapters is build in to the Sun Fire X4100 mother board:
em0 at pci0:1:1:0: class=0x020000 card=0x10118086 chip=0x10108086 rev=0x03
hdr=0x00
    vendor     = 'Intel Corporation'
    device     = '82546EB Gigabit Ethernet Controller (Copper)'
    class      = network
    subclass   = ethernet

TCP_OFFLOAD disabled in kernel's config.

Any ideas?

-- 
MATPOCKuH