[Stgt-devel] Re: [Iscsitarget-devel] a trivial patch to speed up assert a bit

Ming Zhang mingz at ele.uri.edu
Tue Sep 13 04:36:23 CEST 2005


On Tue, 2005-09-13 at 11:30 +0900, FUJITA Tomonori wrote:
> From: Ming Zhang <mingz at ele.uri.edu>
> Subject: Re: [Iscsitarget-devel] a trivial patch to speed up assert a bit
> Date: Mon, 12 Sep 2005 22:16:06 -0400
> 
> > > > > > > > add a unlikely for assert.
> > > > > > > 
> > > > > > > I think that we should kill assert and use BUG_ON.
> > > > > > > 
> > > > > > 
> > > > > > BUG_ON does not have dump_stack()
> > > > > > 
> > > > > > but WARN_ON is weak in some places.
> > > > > 
> > > > > First, simply try to use BUG_ON to learn how it works.
> > > > 
> > > > my dumb.
> > > > 
> > > > [mingz at amd64 iscsitarget-0.4.12]$ grep assert * -rn |wc
> > > >      87     230    3922
> > > > 
> > > > do u have interest to merge the patch if i kill all these assert with
> > > > BUG_ON?
> > > 
> > > To be honest, I don't. There is no chance that IET code is in
> > > mainline, so it is no worth doing that.
> > > 
> > 
> > :) thx. so i will not do that.
> > 
> > so i guess there will not have much development on IET but on stgt then?
> 
> I'll focus on a new iSCSI software target implementation (it is not
> IET, but a fork of IET) for stgt. However, if someone finds serious
> bugs in IET, I'll try to fix them.

Thanks!

ming





More information about the stgt mailing list