linux-kernel - RE: Memory allocation modifications in ibm

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <5ABE8A5E96C0364CAF2B4DC1DEB7CEAC127A2FBE@Mercury.usurf.usu.edu>
Date:	Wed, 1 Sep 2010 21:18:36 +0000
From:	Jonathan Haws <Jonathan.Haws@....usu.edu>
To:	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-net@...r.kernel.org" <linux-net@...r.kernel.org>,
	"linuxppc-dev@...ts.ozlabs.org" <linuxppc-dev@...ts.ozlabs.org>
Subject: RE: Memory allocation modifications in ibm_newemac driver

I found out what was causing the crash, but still am not there and could use some direction:

What was happening was that I was not allocating a new SKB to replace the one in the ring that was being passed up the stack.  I have remedied that and am now having another issue:

Once the ring index rolls over (it does so at 64) I start to lose packets because they are not being handled correctly (or do not contain the correct headers or something of that sort).  Here is a simple ping test showing what is happening:

64 bytes from 172.31.22.21: seq=29 ttl=128 time=10.826 ms
emac/plb/opb/ethernet@...00900: PACKET: 54 0X1800 110 0XC04E6B80
emac/plb/opb/ethernet@...00900: PACKET: 55 0X1800 110 0XC04E6EC0
emac/plb/opb/ethernet@...00900: PACKET: 56 0X1800 98 0XC04E70C0
64 bytes from 172.31.22.21: seq=30 ttl=128 time=10.839 ms
emac/plb/opb/ethernet@...00900: PACKET: 57 0X1800 110 0XC04E6580
emac/plb/opb/ethernet@...00900: PACKET: 58 0X1800 98 0XC04E5B80
64 bytes from 172.31.22.21: seq=31 ttl=128 time=10.832 ms
emac/plb/opb/ethernet@...00900: PACKET: 59 0X1800 219 0XC04E5740
emac/plb/opb/ethernet@...00900: PACKET: 60 0X1800 249 0XC04E5000
emac/plb/opb/ethernet@...00900: PACKET: 61 0X1800 92 0XC04E5340
emac/plb/opb/ethernet@...00900: PACKET: 62 0X1800 98 0XC04E4A00
64 bytes from 172.31.22.21: seq=32 ttl=128 time=10.825 ms
emac/plb/opb/ethernet@...00900: PACKET: 63 0X5800 92 0XC04E4E00
emac/plb/opb/ethernet@...00900: PACKET: 0 0X1800 98 0XC04EF520
emac/plb/opb/ethernet@...00900: PACKET: 1 0X1800 92 0XC04D8340
emac/plb/opb/ethernet@...00900: PACKET: 2 0X1800 98 0XC04D8260
emac/plb/opb/ethernet@...00900: PACKET: 3 0X1800 60 0XC04E7AA0
emac/plb/opb/ethernet@...00900: PACKET: 4 0X1800 92 0XC04E74A0
emac/plb/opb/ethernet@...00900: PACKET: 5 0X1800 92 0XC04E6BC0
emac/plb/opb/ethernet@...00900: PACKET: 6 0X1800 98 0XC04E6F80
emac/plb/opb/ethernet@...00900: PACKET: 7 0X1800 92 0XC04E64E0
emac/plb/opb/ethernet@...00900: PACKET: 8 0X1800 98 0XC04E5BC0

The first number in my debug print statement is what the driver calls the slot number (the ring index).  When it rolls over I start losing the ping replies.  What have I done wrong to cause that?  The data is coming in and is there.  Have any other network device developers seen similar behavior?

Thanks,

Jonathan

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/