Re: [PATCH] Remove if_watchdog use

From: WATANABE Kazuhiro <CQG00620_at_nifty.ne.jp>
Date: Sun, 15 Nov 2009 14:17:57 +0900
Hi,

I've tested the following NICs with your patch on CURRENT, and they
works fine.  Thanks!

* Corega FastEther PCI-TX (DEC 21140-AF)

de0: <Digital 21140A Fast Ethernet> port 0xe000-0xe07f mem 0xd9001000-0xd900107f irq 12 at device 15.0 on pci0
de0: 21140A [10-100Mb/s] pass 2.2
de0: Ethernet address: 00:00:f4:xx:xx:xx
de0: [ITHREAD]

* Acer ALN-201C (Realtek RTL8029AS)

ed0: <RealTek 8029> port 0xe000-0xe01f irq 12 at device 15.0 on pci0
ed0: Ethernet address: 00:60:67:xx:xx:xx
ed0: [ITHREAD]

At Fri, 6 Nov 2009 15:08:22 -0500,
John Baldwin wrote:
> I have a patchset that converts all the remaining users of if_watchdog to 
> using a private callout instead.  In some cases the the driver already used a 
> private timer to drive a stats timer and I merely hooked into that timer.  In 
> other cases a new callout needed to be added to the driver.  Some drivers 
> even abused the if_watchdog interface to provide a stats timer that fired 
> every second. :)  For a few drivers I also fixed other things such as busted 
> locking, order-of-operations issues in detach, or just completely busted 
> drivers (fea(4) and fpa(4) which share the pdq backend).  Please test.  
> Barring any major screaming and shouting I plan to commit this in a week or 
> so and after that to work on removing the if_watchdog/if_timer stuff from the 
> network stack.
> 
> The patch is at http://www.FreeBSD.org/~jhb/patches/cleanup.patch
(snip)
---
WATANABE Kazuhiro (CQG00620_at_nifty.ne.jp)
Received on Sun Nov 15 2009 - 04:18:00 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:58 UTC