libtool-patches
[Top][All Lists]
Advanced

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

Re: Libtool head: stresstest.at export of w8$ symbol


From: Ralf Wildenhues
Subject: Re: Libtool head: stresstest.at export of w8$ symbol
Date: Fri, 2 Sep 2005 16:45:29 +0200
User-agent: Mutt/1.4.1i

Hi Peter,

* Peter Ekberg wrote on Fri, Sep 02, 2005 at 12:03:14PM CEST:
> 
> I'm trying to make sence of the problem with stresstest.at on
> Windows (MinGW, Cygwin and MSVC) where the following symbols
> can't be exported during linking of dlself with -export-symbols
> dlselfsyms:
> 
> w1$
*snip*

Cool!

> I have found a fix which makes it work (not correct I think, but
> attached), but I primarily wonder what the original $ sign is
> doing after the w8 symbol in the dlselfsyms file (the added $
> signs are there to make filtering work, but filtering does not
> happen on Windows). I get the feel that the original $ is
> connected to the first regexp of the sed program at the start
> of the patch context, this first regexp I also don't understand.
> It feels as if there is some way to encode extra (undocumented?)
> information into the symfile. Please tell me what's going on
> here...

I honestly don't have any idea.  I stumbled over a shell parse error
when I looked at the code the last time, did a minimal change to fix
this syntax error, but never understood what this >5 year old code
was supposed to accomplish.

> I think the attached patch is not correct since on other
> platforms the symbols that are not present are filtered out from
> the provided symfile, while on Windows *all* symbols that are
> mentioned in the provided symfile are exported without any
> filtering taking place. Should the symfile be filtered on
> Windows?

Maybe..

> May I also add that it's next to impossible to deduce what's
> actually going on when stresstest.at fails during the umpteenth
> iteration (which happens with this patch applied, when the test
> runs for *much* longer before a new different failure), without
> adding some extra trace echos.

Yes, acknowledged.  I'm very sorry about this, and it's on my list of
things to improve (and I have a couple of ideas), just not with the
highest priority.  If you have a patch ready, feel free to post it. ;)

I'll look at you patch next week, when I come back (unless beaten to)..

Cheers,
Ralf




reply via email to

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