public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* Re: Does cpp still open many files.
       [not found]   ` <200008011521.e71FLi527277@wolery.cumb.org>
@ 2000-08-03  1:16     ` Zack Weinberg
  0 siblings, 0 replies; only message in thread
From: Zack Weinberg @ 2000-08-03  1:16 UTC (permalink / raw)
  To: eliz; +Cc: gcc-bugs

On Tue, Aug 01, 2000 at 06:20:17PM +0200, eliz@is.elta.co.il wrote:
> > Date: Mon, 31 Jul 2000 12:14:01 -0700
> > From: Zack Weinberg <zack@wolery.cumb.org>
> >
> > Setting rlimits such that
> > (number of per-process files allowed) * (number of parallel gcc
> > instances) < (total number of open files supported) should fix that
> > problem.
> 
> I probably don't know enough about the context, but if the call to
> setrlimit is going to be relevant not only to GNU/Linux systems,
> please note that some systems might not have setrlimit in their
> library, or their setrlimit might not support RLIMIT_NOFILE.

I'm suggesting setting rlimits by hand as a temporary workaround, not
a permanent solution wired into the compiler.  (How would we decide
the proper value?)

zw
>From aoliva@redhat.com Thu Aug 03 01:52:00 2000
From: Alexandre Oliva <aoliva@redhat.com>
To: Petr Ovchenkov <ptr@ParaGraph.Ru>
Cc: gcc-bugs@gcc.gnu.org
Subject: Re: g++ unref symbols
Date: Thu, 03 Aug 2000 01:52:00 -0000
Message-id: <or4s52n2u2.fsf@guarana.lsd.ic.unicamp.br>
References: <14728.14999.948376.245409@gargle.gargle.HOWL> <or8zufw915.fsf@guarana.lsd.ic.unicamp.br> <14729.2825.659312.264843@gargle.gargle.HOWL> <orpunqn5pr.fsf@guarana.lsd.ic.unicamp.br> <14729.10392.166467.164794@gargle.gargle.HOWL> <ord7jqn44n.fsf@guarana.lsd.ic.unicamp.br> <14729.12140.47292.22071@gargle.gargle.HOWL>
X-SW-Source: 2000-08/msg00060.html
Content-length: 966

On Aug  3, 2000, Petr Ovchenkov <ptr@ParaGraph.Ru> wrote:

> That's was 2.95.2. For later snapshots (mmm, at least since 20000530)
> for A::A scope instead GLOBAL will be LOCAL (on HP)

I see.  AFAIK, WEAK symbols are now supported and used on HP.  Maybe
they just show up as `t' symbols?  Or maybe it is indeed some new bug?

So, summarizing to the list the discussion we've had in private,
given:

h.h:
struct A {
  A() {}
  virtual void f();
};

x.cc:
#include "h.h"
void A::f() {}

y.cc:
#include "h.h"
A a;

if you compile x.cc and y.cc without optimization, linking fails
because the constructor is defined as local in x.o and not defined at
all in y.  Is that correct?

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist    *Please* write to mailing lists, not to me


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2000-08-03  1:16 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <20000727183822.A4501@lucon.org>
     [not found] ` <20000731121401.J26185@wolery.cumb.org>
     [not found]   ` <200008011521.e71FLi527277@wolery.cumb.org>
2000-08-03  1:16     ` Does cpp still open many files Zack Weinberg

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).