netdev - is PACKET_TX_RING expected to work on loopback?

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>] [day] [month] [year] [list]

Date:	Thu, 6 Oct 2011 14:37:32 -0400
From:	nick black <dankamongmen@...il.com>
To:	Vitali Lovitch <vlovich@...lcomm.com>,
	Johann Baudy <johann.baudy@...-log.net>,
	David Miller <davem@...hat.com>
Cc:	linux-net@...r.kernel.org
Subject: is PACKET_TX_RING expected to work on loopback?

Hey there. I'm developing an application which makes use of PACKET_TX_RING
(http://dank.qemfd.net/dankwiki/index.php/Omphalos), one mmap'd RX and TX
socket per interface. When I transmit on wireless/ethernet devices, the
packets flow out fine, and I get responses.

When I transmit on loopback, I can see the packets using tcpdump -i lo, 
send() returns 0, and the frame is set back to TP_STATUS_AVAILABLE.
Checksums are properly set and the packets appear well-formed in Wireshark.
I can't seem to read the packets, however; neither the intended DNS server
(verified via dnsmasq -d and strace) nor my own socat udp-listen process
seem to get the packets.

Is PACKET_TX_RING expected to work on loopback?

Thanks!

-- 
                                         nick black <nblack@...dia.com>
"A main cause of the Roman Empire's fall was that–lacking zero–they had
       no way to indicate successful termination of their C programs."
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html