Le 14/12/2016 à 08:43, Thomas Huth a écrit : > On 13.12.2016 19:19, Peter Maydell wrote: >> On 9 December 2016 at 12:17, Thomas Huth <th...@redhat.com> wrote: >>> diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c >>> index d4160df..60770d4 100644 >>> --- a/hw/arm/virt-acpi-build.c >>> +++ b/hw/arm/virt-acpi-build.c >>> @@ -33,7 +33,7 @@ >>> #include "qemu/bitmap.h" >>> #include "trace.h" >>> #include "qom/cpu.h" >>> -#include "target-arm/cpu.h" >>> +#include "cpu.h" >>> #include "hw/acpi/acpi-defs.h" >>> #include "hw/acpi/acpi.h" >>> #include "hw/nvram/fw_cfg.h" >> >> Something looks wrong here. We definitely want the ARM >> version of cpu.h, not any random cpu.h. The #include >> filename should make it clear which file we're getting, >> both so it's easier for humans to understand and so that >> one day we might be able to build more than one target >> CPU into the same QEMU binary. > > Right, good catch, thanks! Looks like I did it right in the other > patches and used target/ppc/cpu.h there for example ... I'll fix it for > ARM in the next version of the patch...
There is the same problem in: m68k, alpha, mips and sh4. Could you fix them too? Laurent
