Build errors with Clang 3.2: src/ioport.h:126:18: error: invalid operand for instruction

[Paul Menzel]

Dear SeaBIOS folks,

using Debian Sid/unstable and Clang 3.2 from Debian experimental

$ clang --version
        Debian clang version 3.2-1~exp6 (tags/RELEASE_32/final) (based on LLVM 3.2)
        Target: i386-pc-linux-gnu
        Thread model: posix

to build SeaBIOS, Clang throws the following errors.

$ git show
        commit f465e1ec1b01826100c92b890487a2ab09bfe2c1
        Author: Kevin O'Connor 
        Date:   Sat Feb 23 21:33:36 2013 -0500

Minor - introduce numeric defines for the IVT offset of hw irqs.
        $ make menuconfig # coreboot chosen, see attached
        $ CC=clang make
        […]
        In file included from src/ata.c:9:
        src/ioport.h:126:18: error: invalid operand for instruction
            asm volatile("rep outsw %%es:(%%esi), (%%dx)"
                         ^
        <inline asm>:1:12: note: instantiated into assembly here
                rep outsw %es:(%esi), (%dx)
                          ^~~~~~~~~~
        In file included from src/ata.c:9:
        src/ioport.h:113:18: error: invalid operand for instruction
            asm volatile("rep insw (%%dx), %%es:(%%edi)"
                         ^
        <inline asm>:1:11: note: instantiated into assembly here
                rep insw (%dx), %es:(%edi)
                         ^~~~~
        2 errors generated.
        make: *** [out/ata.o] Fehler 1

I have no clue about assembler programming, so before reporting both
errors to the Clang project, as it works with GCC, I want to make sure
that Clang is mistaken here.

Thanks,

Paul

[Kevin O'Connor]

On Mon, Mar 04, 2013 at 04:15:51PM +0100, Paul Menzel wrote:
> to build SeaBIOS, Clang throws the following errors.

[...]
>     In file included from src/ata.c:9:
>     src/ioport.h:126:18: error: invalid operand for instruction
>         asm volatile("rep outsw %%es:(%%esi), (%%dx)"
>                      ^

I'm not familiar with Clang, but I don't see anything wrong with this
particular assembler statement.  Is this the only thing Clang has
issues with?

-Kevin

[H. Peter Anvin]

On 03/04/2013 03:58 PM, Kevin O'Connor wrote:
> On Mon, Mar 04, 2013 at 04:15:51PM +0100, Paul Menzel wrote:
> > to build SeaBIOS, Clang throws the following errors.
>
> [...]
> >     In file included from src/ata.c:9:
> >     src/ioport.h:126:18: error: invalid operand for instruction
> >         asm volatile("rep outsw %%es:(%%esi), (%%dx)"
> >                      ^
>
> I'm not familiar with Clang, but I don't see anything wrong with this
> particular assembler statement.  Is this the only thing Clang has
> issues with?

Wouldn't simply "es rep outsw" be better?  The operands are implicit
anyway, except for the source segment override.

-hpa

[Kevin O'Connor]

On Tue, Mar 05, 2013 at 03:20:57PM -0800, H. Peter Anvin wrote:
> On 03/04/2013 03:58 PM, Kevin O'Connor wrote:
> > On Mon, Mar 04, 2013 at 04:15:51PM +0100, Paul Menzel wrote:
> > > to build SeaBIOS, Clang throws the following errors.
> >
> > [...]
> > >     In file included from src/ata.c:9:
> > >     src/ioport.h:126:18: error: invalid operand for instruction
> > >         asm volatile("rep outsw %%es:(%%esi), (%%dx)"
> > >                      ^
> >
> > I'm not familiar with Clang, but I don't see anything wrong with this
> > particular assembler statement.  Is this the only thing Clang has
> > issues with?
>
> Wouldn't simply "es rep outsw" be better?  The operands are implicit
> anyway, except for the source segment override.

I kind of prefer to be explicit with the source operands (for
documentation purposes).  Otherwise, it's not really a big deal.
Paul, does Clang prefer "es rep outsw"?

-Kevin

[Paul Menzel]

Am Montag, den 04.03.2013, 18:58 -0500 schrieb Kevin O'Connor:
> On Mon, Mar 04, 2013 at 04:15:51PM +0100, Paul Menzel wrote:
> > to build SeaBIOS, Clang throws the following errors.
>
> [...]
> >     In file included from src/ata.c:9:
> >     src/ioport.h:126:18: error: invalid operand for instruction
> >         asm volatile("rep outsw %%es:(%%esi), (%%dx)"
> >                      ^
>
> I'm not familiar with Clang, but I don't see anything wrong with this
> particular assembler statement.

Alright, then I will report that to the Clang folks.
> Is this the only thing Clang has issues with?

I am not sure, as the build stops after this error.

Thanks,

Paul

[Paul Menzel]

Am Mittwoch, den 06.03.2013, 11:49 +0100 schrieb Paul Menzel:
> Am Montag, den 04.03.2013, 18:58 -0500 schrieb Kevin O'Connor:
> > On Mon, Mar 04, 2013 at 04:15:51PM +0100, Paul Menzel wrote:
> > > to build SeaBIOS, Clang throws the following errors.
> >
> > [...]
> > >     In file included from src/ata.c:9:
> > >     src/ioport.h:126:18: error: invalid operand for instruction
> > >         asm volatile("rep outsw %%es:(%%esi), (%%dx)"
> > >                      ^
> >
> > I'm not familiar with Clang, but I don't see anything wrong with this
> > particular assembler statement.
>
> Alright, then I will report that to the Clang folks.

This issue is tracked in ticket 15455 in the LLVM Bugzilla bug tracking
system [1].
> > Is this the only thing Clang has issues with?
>
> I am not sure, as the build stops after this error.

[…]

Thanks,

Paul

[1] http://llvm.org/bugs/show_bug.cgi?id=15455