xfs
[Top] [All Lists]

Re: do_softirq

To: Keith Owens <kaos@xxxxxxxxxxxxxxxxx>
Subject: Re: do_softirq
From: Walt H <waltabbyh@xxxxxxxxxx>
Date: Thu, 14 Jun 2001 19:18:37 -0700
Cc: "Bernhard R. Erdmann" <be@xxxxxxxxxxx>, Linux XFS Mailing List <linux-xfs@xxxxxxxxxxx>
References: <2343.992570076@kao2.melbourne.sgi.com>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux 2.4.5-xfs i686; en-US; rv:0.9.1+) Gecko/20010614
Thanks Keith.

I'm an experienced linux user, but a crappy kernel programmer :) Always good to have clarity.


OT - Just ready about "The Search for The Holy Grail" on Slashdot. Keep wanting to say: "Behind the rabbit?" Sorry... I'll go away now.


-Walt



Keith Owens wrote:

On Thu, 14 Jun 2001 18:42:26 -0700, Walt H <waltabbyh@xxxxxxxxxx> wrote:

This is a different patch than the one I was aware of from the kml. Is
this preferred over the non-version symbol export? (Hope I didn't misunderstand that..) That one seemed to work well.



Both patches are mine. It required about 10 emails between various kernel hackers and digging into the more obscure bits of gcc to agree on a method of exposing calls to exported symbols from __asm__ code. The change to softirq.h is the correct mechanism, especially as you do not have to run make mrproper after applying the patch, changing from EXPORT_SYMBOL() to EXPORT_SYMBOL_NOVERS() requires make mrproper.





<Prev in Thread] Current Thread [Next in Thread>