netdev - [PATCH 0/3]: v2 of "TX backlog management in-driver..."

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-next>] [day] [month] [year] [list]

Message-Id: <20080620.161634.193691261.davem@davemloft.net>
Date:	Fri, 20 Jun 2008 16:16:34 -0700 (PDT)
From:	David Miller <davem@...emloft.net>
To:	netdev@...r.kernel.org
CC:	vinay@...ux.vnet.ibm.com, krkumar2@...ibm.com, mchan@...adcom.com
Subject: [PATCH 0/3]: v2 of "TX backlog management in-driver..."


Changes since first version:

1) Integrate TSO fixes and cleanups from Michael Chan into
   patch one.  Michael if I could have your signoff on that
   one, I'd appreciate it.

2) Patch three attempts to deal with the reordering issue and
   races as follows:

   a) ->hard_start_xmit() considers the TX ring full if any
      packets are in the TX backlog queue.  This ensures that
      the backlog is fully empties before fresh ->hard_start_xmit()
      packets can be put onto the TX ring

   b) Mimicking the existing netif_{stop,wake}_queue() machinery,
      tg3_tx() and ->hard_start_xmit() cooperate to make updates,
      tests, and memory barriers such that the the TX backlog will
      not get wedged.  There are extensive comments added to the
      code in this patch which tries to explain how this works.

Patch two is essentially unchanged.

Signed-off-by: David S. Miller <davem@...emloft.net>
--
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