[Bug target/65464] New: [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[Bug target/65464] New: [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

            Bug ID: 65464
           Summary: [4.8/4.9 Regression] disabling multilib support for
                    powerpc64le-linux-gnu breaks kernel builds
           Product: gcc
           Version: 4.9.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: doko at gcc dot gnu.org

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

Matthias Klose <doko at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |powerpc64le-linux-gnu
                 CC|                            |amodra at gcc dot gnu.org,
                   |                            |msebor at gcc dot gnu.org
      Known to work|                            |4.8.4, 4.9.2
      Known to fail|                            |4.8.5, 4.9.3

--- Comment #1 from Matthias Klose <doko at gcc dot gnu.org> ---
removing the multilib support on powerpc64le-linux-gnu just to be able to omit
a configure option doesn't seem to appropriate.  At least it breaks older linux
kernel builds, which then decide to build the vdso32 stuff just without -m32,
and fails.  Please revert this change on the branches.

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1433809

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I think this is a bug in the kernel.  vdso32 should be building for
little-endian PowerPC64 as there is no 32bit support currently.

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[amodra at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

Alan Modra <amodra at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |amodra at gmail dot com
         Resolution|---                         |WONTFIX

--- Comment #3 from Alan Modra <amodra at gmail dot com> ---
32-bit support for powerpc64le hasn't been removed, it is just not enabled by
default since we don't support a 32-bit user runtime.  As Andrew said, it is a
kernel bug that a 32-bit vdso is built.

You can get 32-bit compiler support by configuring with
--enable-targets=powerpcle-linux --disable-multilib
if you so desire.

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[trippels at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

Markus Trippelsdorf <trippels at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |trippels at gcc dot gnu.org

--- Comment #4 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
Well, on x86_64 if you build gcc with --disable-multilib you still
can compile with -m32 (even without a 32-bit user runtime).
Why should this be different on ppc64le?

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[doko at ubuntu dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #5 from Matthias Klose <doko at ubuntu dot com> ---
> Well, on x86_64 if you build gcc with --disable-multilib you still
> can compile with -m32 (even without a 32-bit user runtime).
> Why should this be different on ppc64le?

$ gcc -m32 -c foo.c
foo.c:1:0: error: -m32 not supported in this configuration

well, it is different, and this doesn't work on 4.8, 4.9 and 5 anymore.

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[trippels at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #6 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
(In reply to Matthias Klose from comment #5)
> > Well, on x86_64 if you build gcc with --disable-multilib you still
> > can compile with -m32 (even without a 32-bit user runtime).
> > Why should this be different on ppc64le?
> 
> $ gcc -m32 -c foo.c
> foo.c:1:0: error: -m32 not supported in this configuration
> 
> well, it is different, and this doesn't work on 4.8, 4.9 and 5 anymore.

I agree with you. The question was directed to Alan.

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[amodra at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #7 from Alan Modra <amodra at gmail dot com> ---
On x86_64-linux you can run 32-bit binaries.  On powerpc64le-linux you can't. 
It's that simple.  -m32 support in powerpc64le gcc, by default, doesn't make
sense until we have
- supported and tested compat layer in the kernel
- supported and tested gcc, binutils and glibc
We might even want to change the 32-bit ABI for powerpcle.

The change I made wasn't "just to be able to omit" --disable-multilib.  It was
also to fix the wrong default of enabling -m32 support in powerpc64le gcc.

Another data point: llvm doesn't support -m32

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #8 from Matthias Klose <doko at gcc dot gnu.org> ---
Alan told me that you can work around it by configuring with

  --enable-targets=powerpcle-linux --disable-multilib

to still have the -m32 option recognized.

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[trippels at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #9 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
(In reply to Alan Modra from comment #7)
> On x86_64-linux you can run 32-bit binaries.  On powerpc64le-linux you
> can't.  It's that simple.  -m32 support in powerpc64le gcc, by default,
> doesn't make sense until we have
> - supported and tested compat layer in the kernel
> - supported and tested gcc, binutils and glibc

Only if IA32_EMULATION is set in the kernel config. I have disabled
this option for years now. So the situation is exactly the same.

[Bug target/65464] [4.8/4.9 Regression] disabling multilib support for powerpc64le-linux-gnu breaks kernel builds

[trippels at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65464

--- Comment #10 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
And even with !IA32_EMULATION the kernel build uses -m16 a few times on x86_64.
Please note that we're talking about -ffreestanding target here.