On 1/25/21 5:29 PM, Thomas Huth wrote:
> On 25/01/2021 16.37, Philippe Mathieu-Daudé wrote:
>> It is pointless to build the s390-ccw BIOS when only user-mode
>> emulation is built. Only build it when s390 system mode emulation
>> is selected.
>>
>> Signed-off-by: Philippe Mathieu-Daudé <f4...@amsat.org>
>> ---
>> v2: Restrict to s390x-softmmu (thuth)
>>
>> We could move the whole block into the case(), but
>> 1/ it doesn't look nice
>> 2/ it doesn't match the rest of this file code style
>>
>> Supersedes: <20210124161740.2502786-1-f4...@amsat.org>
>> ---
>> configure | 7 ++++++-
>> 1 file changed, 6 insertions(+), 1 deletion(-)
>>
>> diff --git a/configure b/configure
>> index dcc5ea7d630..997fb0761a5 100755
>> --- a/configure
>> +++ b/configure
>> @@ -461,6 +461,7 @@ skip_meson=no
>> gettext="auto"
>> fuse="auto"
>> fuse_lseek="auto"
>> +s390_ccw_bios="no"
>> malloc_trim="auto"
>> @@ -5373,7 +5374,11 @@ if { test "$cpu" = "i386" || test "$cpu" =
>> "x86_64"; } && \
>> fi
>> # Only build s390-ccw bios if we're on s390x and the compiler has
>> -march=z900
>> -if test "$cpu" = "s390x" ; then
>> +case $target_list in
>> + *"s390x-softmmu"*) s390_ccw_bios=yes
>> + ;;
>> +esac
>> +if test "$s390_ccw_bios" = yes; then
>
> Why did you drop the test for the host CPU? ... the s390-ccw bios
> currently only compiles on a s390x box...
I saw something odd but couldn't figure. Too much squashing and
switching between branches...
Phil.
