Re: [Qemu-devel] [RFC PATCH 10/25] build: convert pci.mak to Kconfig

[Yang Zhong <yang.zhong@intel.com>]

On Fri, Jan 04, 2019 at 02:48:03PM +0100, Thomas Huth wrote:
> On 2018-12-27 07:34, Yang Zhong wrote:
> > From: Paolo Bonzini <pbonzini@redhat.com>
> > 
> > Instead of including the same list of devices for each target,
> > set CONFIG_PCI to true, and make the devices default to present
> > whenever PCI is available.
> > 
> > Done mostly with the following script:
> > 
> >   while read i; do
> >      i=${i%=y}; i=${i#CONFIG_}
> >      sed -i -e'/^config '$i'$/!b' -en \
> >             -e'a\' -e'    default y\' -e'    depends on PCI' \
> >           `grep -lw $i hw/*/Kconfig`
> >   done < default-configs/pci.mak
> > 
> > followed by replacing a few "depends on" clauses with "select"
> > whenever the symbol is not really related to PCI.
> > 
> > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> > Signed-off-by: Yang Zhong <yang.zhong@intel.com>
> > ---
> >  default-configs/i386-softmmu.mak |  2 +-
> >  default-configs/pci.mak          | 47 --------------------------------
> >  hw/audio/Kconfig                 |  6 ++++
> >  hw/block/Kconfig                 |  2 ++
> >  hw/char/Kconfig                  |  2 ++
> >  hw/display/Kconfig               | 12 +++++++-
> >  hw/ide/Kconfig                   |  6 ++++
> >  hw/ipack/Kconfig                 |  2 ++
> >  hw/misc/Kconfig                  |  4 +++
> >  hw/net/Kconfig                   | 19 +++++++++++++
> >  hw/scsi/Kconfig                  | 11 ++++++++
> >  hw/sd/Kconfig                    |  3 ++
> >  hw/usb/Kconfig                   | 10 +++++++
> >  hw/virtio/Kconfig                |  3 ++
> >  hw/watchdog/Kconfig              |  2 ++
> >  15 files changed, 82 insertions(+), 49 deletions(-)
> >  delete mode 100644 default-configs/pci.mak
> [...]
> > diff --git a/hw/char/Kconfig b/hw/char/Kconfig
> > index 26c13243cf..1ed6f0dbce 100644
> > --- a/hw/char/Kconfig
> > +++ b/hw/char/Kconfig
> > @@ -15,6 +15,8 @@ config SERIAL_ISA
> >  
> >  config SERIAL_PCI
> >      bool
> > +    default y
> > +    depends on PCI
> 
> I think this likely needs a "select SERIAL", too? At least the
> SERIAL_ISA switch gets a "select SERIAL" in patch 15/25, so it sounds
> logical if the SERIAL_PCI would get this, too...
> 
>  Thomas

  Thomas, thanks, i will add "select SERIAL" into 15/25 patch, thanks again!

  Regards,

  Yang