[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: faster AS_IF,AS_CASE

From: Ralf Wildenhues
Subject: Re: faster AS_IF,AS_CASE
Date: Thu, 14 Aug 2008 20:39:25 +0200
User-agent: Mutt/1.5.18 (2008-05-17)

Hi Eric,

* Eric Blake wrote on Wed, Aug 13, 2008 at 06:27:26PM CEST:
> I'm not applying this patch on savannah yet, in case there are any
> comments.  

This patch looks good to me, thank you.  It also seems to not incur any
measureable slowdown in the common case of few arguments.  I've tried it
on OpenMPI which uses AS_IF a lot.

And yes, I have changed my mind about this from what I wrote earlier
(off-list).  The factorization is very nice indeed, the slowdown is
apparently in the noise range, and the change looks safer than I feared.

> For example, is my new macro m4_foreach_pair worth documenting as a supported 
> interface, or should I leave it undocumented?

How about documenting it after 2.63, so we can change it until the next
version if we find issues with it?

> Also, should it be named slightly differently since it behaves more
> like m4_map than m4_foreach?

m4_map_pair maybe?

Cheers, and thanks for all your work on this,

> From: Eric Blake <address@hidden>
> Date: Wed, 13 Aug 2008 09:12:28 -0600
> Subject: [PATCH] Implement m4_foreach_pair.
> * lib/m4sugar/m4sugar.m4 (m4_foreach_pair): New macro,
> undocumented for now.
> * lib/m4sugar/foreach.m4 (m4_foreach_pair): Also the m4 1.4.x
> counterpart.
> * lib/m4sugar/m4sh.m4 (AS_IF, AS_CASE): Use it.
> * tests/ (AS@&address@hidden and AS@&address@hidden): Test it.
> Signed-off-by: Eric Blake <address@hidden>

Signed-off-by: Ralf Wildenhues <address@hidden>

reply via email to

[Prev in Thread] Current Thread [Next in Thread]