This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: gcc 3.1 fails to build Linux 2.4.7 kernel on i686-linux

It works. I think a good gcc and a "proper" kernel are converging :)

gcc 3.1 pre-release (20020311 snapshot) successfully built a bootable 2.4.18
kernel. The preemption patch and lock-break patch for 2.4.18 were applied and
configured before the build. All 2.4.18 kernel modules needed by my system
were built and installed. The distribution is x86 RedHat 7.2 running on a
uniprocessor Athlon box.

----- Original Message -----
From: "Craig Rodrigues" <>
To: <>
Sent: Sunday, March 17, 2002 12:18 AM
Subject: Re: gcc 3.1 fails to build Linux 2.4.7 kernel on i686-linux

> On Sat, Mar 16, 2002 at 10:13:40PM +0000, Joseph S. Myers wrote:
> >
> > In practice this part of the release criteria didn't seem to get applied
> > for 3.0 - there weren't any regular status reports before the release
> > collecting details of what worked on what platforms, and what needed to be
> > made to work before release, and Linux 2.4.x compilation with 3.0.y didn't
>                                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > work until about 3.0.3.
> ^^^^^^^^^^^^^^^^^^^^^^^^^
> Joseph is absolutely right in this statement, and the fact that
> the GCC 3.0 release could go out the door, and not be able to
> compile an important piece of software like the Linux kernel until GCC 3.0.3
> should be considered an embarassment to the GCC project.
> I saw many, many bug reports in GNATS from people who failed to
> compile the Linux kernel with GCC 3.0.x, where x < 3.
> We need to get our act together and figure what can and cannot
> work in terms of GCC compiling the Linux kernel.
> If you have free time, disk space, and CPU, I urge you to help
> join in the GCC testing effort before the GCC 3.1 release. :)
> I've heard moaning and groaning in the past about how the Linux
> kernel takes advantage of implementation details of various versions
> of GCC.  Well, this is the real world of kernel development, and
> we have to live in the real world, where real users are going to
> want to compile their kernel with gcc.  That's life.
> If the Linux kernel needs to be fixed, we should notify the Linux kernel
> people with the appropriate fixes.  If gcc needs to be fixed, we should fix
> If the gcc release criteria says that compiling Linux kernel 2.4.0 with
> gcc 3.1 is OK when it is not, then the release criteria should be fixed
> to tell the truth.
> I just grabbed Linux kernel 2.4.18 from the Red Hat rawhide ftp site,
> so I'll give that a whirl and see how it goes.
> --
> Craig Rodrigues

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]