netdev
[Top] [All Lists]

8139C+/8169 and suspend mode

To: netdev@xxxxxxxxxxx
Subject: 8139C+/8169 and suspend mode
From: Francois Romieu <romieu@xxxxxxxxxxxxx>
Date: Sat, 25 Sep 2004 21:40:01 +0200
Cc: jgarzik@xxxxxxxxx, Jon Mason <jdmason@xxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4.1i
Does anyone have positive experience with suspend mode on the aforementionned
chipset ?

Rationale: Jon noticed that the r8169 driver did not correctly set the dirty
Rx ring index when the driver tries to reset the chipset (rtl8169_hw_start)
after a Tx timeout recovery. The chipset is told where the Tx/Rx rings start
but the software driver works with a badly inaccurate (rx_cur, rx_dirty) pair.

If I am not mistaken, the same pattern applies to the resume function in the
r8169 driver and in the 8139cp driver. 

So, despite me thinking that the poor thing is in a bad state, is there
anybody who actually succeeds using it ?

--
Ueimor

<Prev in Thread] Current Thread [Next in Thread>
  • 8139C+/8169 and suspend mode, Francois Romieu <=