Add the header and its dependencies.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
 scripts/update-linux-headers.sh | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/scripts/update-linux-headers.sh b/scripts/update-linux-headers.sh
index d18e2f1..e528bda 100755
--- a/scripts/update-linux-headers.sh
+++ b/scripts/update-linux-headers.sh
@@ -80,11 +80,6 @@ for arch in $ARCHLIST; do
         continue
     fi
 
-    # Blacklist architectures which have KVM headers but are actually dead
-    if [ "$arch" = "ia64" -o "$arch" = "mips" ]; then
-        continue
-    fi
-
     if [ "$arch" = x86 ]; then
         arch_var=SRCARCH
     else
@@ -95,9 +90,17 @@ for arch in $ARCHLIST; do
 
     rm -rf "$output/linux-headers/asm-$arch"
     mkdir -p "$output/linux-headers/asm-$arch"
-    for header in kvm.h kvm_para.h unistd.h; do
+    for header in unistd.h bitsperlong.h; do
         cp "$tmpdir/include/asm/$header" "$output/linux-headers/asm-$arch"
     done
+
+    # Blacklist architectures which have KVM headers but are actually dead
+    if [ "$arch" != "ia64" -a "$arch" != "mips" ]; then
+        for header in kvm.h kvm_para.h; do
+            cp "$tmpdir/include/asm/$header" "$output/linux-headers/asm-$arch"
+        done
+    fi
+
     if [ $arch = powerpc ]; then
         cp "$tmpdir/include/asm/epapr_hcalls.h" "$output/linux-headers/asm-powerpc/"
     fi
@@ -130,7 +133,7 @@ for header in kvm.h kvm_para.h vfio.h vfio_ccw.h vhost.h \
 done
 rm -rf "$output/linux-headers/asm-generic"
 mkdir -p "$output/linux-headers/asm-generic"
-for header in kvm_para.h; do
+for header in kvm_para.h bitsperlong.h unistd.h; do
     cp "$tmpdir/include/asm-generic/$header" "$output/linux-headers/asm-generic"
 done
 if [ -L "$linux/source" ]; then
-- 
MST

On 20 March 2018 at 17:17, Michael S. Tsirkin <mst@redhat.com> wrote:
> Add the header and its dependencies.
>
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> ---
>  scripts/update-linux-headers.sh | 17 ++++++++++-------
>  1 file changed, 10 insertions(+), 7 deletions(-)
>
> diff --git a/scripts/update-linux-headers.sh b/scripts/update-linux-headers.sh
> index d18e2f1..e528bda 100755
> --- a/scripts/update-linux-headers.sh
> +++ b/scripts/update-linux-headers.sh
> @@ -80,11 +80,6 @@ for arch in $ARCHLIST; do
>          continue
>      fi
>
> -    # Blacklist architectures which have KVM headers but are actually dead
> -    if [ "$arch" = "ia64" -o "$arch" = "mips" ]; then
> -        continue
> -    fi
> -
>      if [ "$arch" = x86 ]; then
>          arch_var=SRCARCH
>      else
> @@ -95,9 +90,17 @@ for arch in $ARCHLIST; do
>
>      rm -rf "$output/linux-headers/asm-$arch"
>      mkdir -p "$output/linux-headers/asm-$arch"
> -    for header in kvm.h kvm_para.h unistd.h; do
> +    for header in unistd.h bitsperlong.h; do
>          cp "$tmpdir/include/asm/$header" "$output/linux-headers/asm-$arch"
>      done
> +
> +    # Blacklist architectures which have KVM headers but are actually dead
> +    if [ "$arch" != "ia64" -a "$arch" != "mips" ]; then

We don't need to blacklist ia64 here, beacues it doesn't have the kvm
headers in the kernel any more, so it is already being skipped because
of the condition at the top of the loop that ignores any arch which
doesn't have a kvm.h.

For MIPS, KVM is supposed to be properly supported these days, right?
That sounds like we should not be blacklisting them either, but instead
sorting out whatever issues it was that made us exclude them.

> +        for header in kvm.h kvm_para.h; do
> +            cp "$tmpdir/include/asm/$header" "$output/linux-headers/asm-$arch"
> +        done
> +    fi
> +
>      if [ $arch = powerpc ]; then
>          cp "$tmpdir/include/asm/epapr_hcalls.h" "$output/linux-headers/asm-powerpc/"
>      fi
> @@ -130,7 +133,7 @@ for header in kvm.h kvm_para.h vfio.h vfio_ccw.h vhost.h \
>  done
>  rm -rf "$output/linux-headers/asm-generic"
>  mkdir -p "$output/linux-headers/asm-generic"
> -for header in kvm_para.h; do
> +for header in kvm_para.h bitsperlong.h unistd.h; do
>      cp "$tmpdir/include/asm-generic/$header" "$output/linux-headers/asm-generic"
>  done
>  if [ -L "$linux/source" ]; then
> --
> MST

thanks
-- PMM

On Tue, Mar 20, 2018 at 05:24:20PM +0000, Peter Maydell wrote:
> On 20 March 2018 at 17:17, Michael S. Tsirkin <mst@redhat.com> wrote:
> > Add the header and its dependencies.
> >
> > Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> > ---
> >  scripts/update-linux-headers.sh | 17 ++++++++++-------
> >  1 file changed, 10 insertions(+), 7 deletions(-)
> >
> > diff --git a/scripts/update-linux-headers.sh b/scripts/update-linux-headers.sh
> > index d18e2f1..e528bda 100755
> > --- a/scripts/update-linux-headers.sh
> > +++ b/scripts/update-linux-headers.sh
> > @@ -80,11 +80,6 @@ for arch in $ARCHLIST; do
> >          continue
> >      fi
> >
> > -    # Blacklist architectures which have KVM headers but are actually dead
> > -    if [ "$arch" = "ia64" -o "$arch" = "mips" ]; then
> > -        continue
> > -    fi
> > -
> >      if [ "$arch" = x86 ]; then
> >          arch_var=SRCARCH
> >      else
> > @@ -95,9 +90,17 @@ for arch in $ARCHLIST; do
> >
> >      rm -rf "$output/linux-headers/asm-$arch"
> >      mkdir -p "$output/linux-headers/asm-$arch"
> > -    for header in kvm.h kvm_para.h unistd.h; do
> > +    for header in unistd.h bitsperlong.h; do
> >          cp "$tmpdir/include/asm/$header" "$output/linux-headers/asm-$arch"
> >      done
> > +
> > +    # Blacklist architectures which have KVM headers but are actually dead
> > +    if [ "$arch" != "ia64" -a "$arch" != "mips" ]; then
> 
> We don't need to blacklist ia64 here, beacues it doesn't have the kvm
> headers in the kernel any more, so it is already being skipped because
> of the condition at the top of the loop that ignores any arch which
> doesn't have a kvm.h.

OK.

> For MIPS, KVM is supposed to be properly supported these days, right?
> That sounds like we should not be blacklisting them either, but instead
> sorting out whatever issues it was that made us exclude them.

I agree but I'd rather someone else looked at this.

> > +        for header in kvm.h kvm_para.h; do
> > +            cp "$tmpdir/include/asm/$header" "$output/linux-headers/asm-$arch"
> > +        done
> > +    fi
> > +
> >      if [ $arch = powerpc ]; then
> >          cp "$tmpdir/include/asm/epapr_hcalls.h" "$output/linux-headers/asm-powerpc/"
> >      fi
> > @@ -130,7 +133,7 @@ for header in kvm.h kvm_para.h vfio.h vfio_ccw.h vhost.h \
> >  done
> >  rm -rf "$output/linux-headers/asm-generic"
> >  mkdir -p "$output/linux-headers/asm-generic"
> > -for header in kvm_para.h; do
> > +for header in kvm_para.h bitsperlong.h unistd.h; do
> >      cp "$tmpdir/include/asm-generic/$header" "$output/linux-headers/asm-generic"
> >  done
> >  if [ -L "$linux/source" ]; then
> > --
> > MST
> 
> thanks
> -- PMM

On 20/03/2018 18:29, Michael S. Tsirkin wrote:
>> For MIPS, KVM is supposed to be properly supported these days, right?
>> That sounds like we should not be blacklisting them either, but instead
>> sorting out whatever issues it was that made us exclude them.
>
> I agree but I'd rather someone else looked at this.

Yeah, this doesn't look like the best time to change that.

Thanks,

Paolo