All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pete Popov <ppopov@pacbell.net>
To: Ralf Baechle <ralf@oss.sgi.com>
Cc: Keith Owens <kaos@melbourne.sgi.com>,
	linux-mips@oss.sgi.com, linux-mips@fnet.fr
Subject: Re: Kernel compiler
Date: Sat, 04 Nov 2000 03:10:50 +0000 (GMT)	[thread overview]
Message-ID: <20001104.3105000@zeus.mvista.com> (raw)
In-Reply-To: <20001104035326.A29005@bacchus.dhis.org>



>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 11/3/00, 6:53:26 PM, Ralf Baechle <ralf@oss.sgi.com> wrote regarding Re: 
Kernel compiler:


> On Sat, Nov 04, 2000 at 01:44:22PM +1100, Keith Owens wrote:

> > On Fri, 3 Nov 2000 22:19:26 +0100,
> > Ralf Baechle <ralf@oss.sgi.com> wrote:
> > >Due to compiler bugs with named initializers the use of egcs 1.1.2 has
> > >ben deprecated for Linux; Linux 2.4.0-test10 will refuse to compile with
> > >older compilers.
> >
> > Correction.  gcc 2.7 has been deprecated, the recommended version is
> > gcc 2.91.66 (also known as egcs 1.1.2) or better.  Unfortunately RedHat
> > created an unofficial 2.96 which is known to miscompile the kernel, so
> > you cannot just use the "latest" gcc.  It is believed that gcc versions
> > 2.91.66 through 2.95 inclusive are safe, as long as you use
> > -fno-strict-aliasing.

> Sigh, yes you're right as other have already pointed out on IRC ...

> Gcc 2.7 is no longer being used for Linux/MIPS since a long time and I'd
> simply bitbucket any bugreport related to it.  Most people are using egcs
> 1.0.3a for everything which is older than the required minimum version 
1.1.2,
> so we're in some trouble.

> I've got reports regarding the 1.1.2 crosscompiler which say that it
> misscompiles MIPS kernels, sigh ...

What are the symptoms of a "miscompiled" kernel and how does one detect 
that (assuming that the kernel actually compiles successfully, but has 
problems running).

Pete

  parent reply	other threads:[~2000-11-04  3:10 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-11-03 21:19 Kernel compiler Ralf Baechle
2000-11-04  2:44 ` Keith Owens
2000-11-04  2:53   ` Ralf Baechle
2000-11-04  3:03     ` Keith M Wesolowski
2000-11-04  3:10     ` Pete Popov [this message]
2000-11-04  3:20       ` Ralf Baechle
2000-11-04  3:13     ` Keith Owens
2000-11-04 10:40     ` Alan Cox
2000-11-04 10:40       ` Alan Cox
2000-11-07  3:39       ` Ralf Baechle
  -- strict thread matches above, loose matches on Subject: below --
2001-10-25 14:31 kernel compiler Frontgate Lab
2001-10-25 15:08 ` David Weinehall
2001-10-25 15:28 ` Alan Cox
2001-10-25 15:27   ` Steve Lord
2001-10-25 15:46     ` Jakub Jelinek
2001-10-25 15:45       ` Steve Lord
2001-10-25 15:20 John Weber
2001-10-25 15:26 ` Axel
2001-10-25 15:45 ` Alan Cox
2001-10-25 16:45   ` Lost Logic
2001-10-25 17:16     ` Alan Cox
2001-10-25 22:00       ` J . A . Magallon
2001-10-25 22:13     ` J . A . Magallon
2001-10-26  9:18       ` Allan Sandfeld
2001-10-26  9:35         ` Jakub Jelinek

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20001104.3105000@zeus.mvista.com \
    --to=ppopov@pacbell.net \
    --cc=kaos@melbourne.sgi.com \
    --cc=linux-mips@fnet.fr \
    --cc=linux-mips@oss.sgi.com \
    --cc=ralf@oss.sgi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.