Re: [Qemu-devel] [PULL v2 00/54] s390x update

[Cornelia Huck]

On Tue, 21 May 2019 14:24:07 +0100
Peter Maydell <address@hidden> wrote:

> On Tue, 21 May 2019 at 10:20, Cornelia Huck <address@hidden> wrote:
> >
> > [Note: I'm only sending the changed headers update to spare everybody 
> > another
> > patch avalanche. The remainder of the patches is unchanged, save for the
> > changed commit ids of the cpumodel patches on top of the headers update.]
> >
> > The following changes since commit d8276573da58e8ce78dab8c46dd660efd664bcb7:
> >
> >   Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-20190510' into 
> > staging (2019-05-16 13:15:08 +0100)
> >
> > are available in the Git repository at:
> >
> >   https://github.com/cohuck/qemu tags/s390x-20190521-2
> >
> > for you to fetch changes up to 4d727d1aa0e5cbf9f5f00968698dfa34c7b47b08:
> >
> >   s390x/cpumodel: wire up 8561 and 8562 as gen15 machines (2019-05-21 
> > 10:40:03 +0200)
> >
> > ----------------------------------------------------------------
> > s390x update:
> > - have the bios tolerate bootmap signature entries
> > - next chunk of vector instruction support in tcg
> > - a headers update against Linux 5.2-rc1
> > - add more facilities and gen15 machines to the cpu model  
> 
> Hi -- looks like the update-linux-headers change isn't quite right:
> 
> --- a/scripts/update-linux-headers.sh
> +++ b/scripts/update-linux-headers.sh
> @@ -122,6 +122,9 @@ for arch in $ARCHLIST; do
>          cp "$tmpdir/include/asm/unistd-oabi.h" 
> "$output/linux-headers/asm-arm/"
>          cp "$tmpdir/include/asm/unistd-common.h"
> "$output/linux-headers/asm-arm/"
>      fi
> +    if [ $arch = arm64 ]; then
> +        cp "$tmpdir/include/asm/sve_context.h" 
> "$output/linux-headers/asm-arm/"
> +    fi
>      if [ $arch = x86 ]; then
>          cp "$tmpdir/include/asm/unistd_32.h" "$output/linux-headers/asm-x86/"
>          cp "$tmpdir/include/asm/unistd_x32.h" 
> "$output/linux-headers/asm-x86/"
> 
> ...it's copying the file into asm-arm/ rather than asm-arm64/.
> (I did a by-hand move of the file into the right directory and
> that was sufficient for the compile to succeed.)

I'm obviously not at my best at the moment :(

> 
> Also, can we keep the changes to scripts/ in a separate
> commit, please? The idea of header-update commits is that they
> should contain only the changes automatically generated
> by the script, with no hand-written additions.

This looks a bit like a chicken-and-egg problem, though... without the
change, we cannot point the script at a current kernel tree. I'd prefer
the two to stay together, but I can certainly point out the change to
the script more prominently.