On Mon, Jul 04, 2022 at 09:38:46AM -0600, Warner Losh wrote:
> On Mon, Jul 4, 2022 at 9:28 AM Daniel P. Berrangé <berrange@redhat.com>
> wrote:
>
> > A few files relied on qemu/osdep.h being included via a common
> > header. Another file didn't need it because it was actually an
> > included file, so ought to have been named .c.inc
> >
> > Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
> > ---
> > bsd-user/arm/signal.c | 2 ++
> > bsd-user/arm/target_arch_cpu.c | 3 +++
> > bsd-user/{elfcore.c => elfcore.c.inc} | 0
> > bsd-user/elfload.c | 2 +-
> > bsd-user/freebsd/os-sys.c | 2 ++
> > bsd-user/i386/signal.c | 2 ++
> > bsd-user/qemu.h | 1 -
> > bsd-user/x86_64/signal.c | 2 ++
> > crypto/rsakey.c | 1 +
> > qga/cutils.c | 2 ++
> > 10 files changed, 15 insertions(+), 2 deletions(-)
> > rename bsd-user/{elfcore.c => elfcore.c.inc} (100%)
> >
>
> The change to bsd-user is fine, though will cause many ripples in the
> upstream
> branch when I merge it. The ripples likely are worth it in the long run,
> and knowing
> they are coming and helps me prepare the tree for the merge.
If you prefer to delay these changes I don't mind. It just means that
it would need a 'bsd-user/.*' exclude rule in the next patch to
temporarily skip this chck for bsd-user code.
No. Go ahead. I know I need to do something when I do the next merge.
And it shouldn't be a big deal to do now, otherwise I'll forget...
Warner