On 04/29/2014 03:10 AM, Chunyan Liu wrote: > Add def_value_str (default value) to QemuOptDesc, to replace function of the > default value in QEMUOptionParameter. > > Improve qemu_opts_get_* functions: if find opt, return opt->str; otherwise, > if desc->def_value_str is set, return desc->def_value_str; otherwise, return > input defval. > > Improve qemu_opts_print: if option is set, print opt->str; otherwise, if > desc->def_value_str is set, also print it. It will replace > print_option_parameters. > > Signed-off-by: Dong Xu Wang <wdon...@linux.vnet.ibm.com> > Signed-off-by: Chunyan Liu <cy...@suse.com> > --- > Changes to V25: > * split v25 patch into two: this one and next one. > * this patch is the same as v22 which is reviewed-by Eric. > > include/qemu/option.h | 3 ++- > util/qemu-option.c | 56 > ++++++++++++++++++++++++++++++++++++++++++--------- > 2 files changed, 49 insertions(+), 10 deletions(-)
> @@ -895,17 +918,32 @@ void qemu_opts_del(QemuOpts *opts)
> g_free(opts);
> }
>
> -int qemu_opts_print(QemuOpts *opts, void *dummy)
> +void qemu_opts_print(QemuOpts *opts)
> {
This part of the hunk belongs with the changes in 3/32. I think you got
the split backwards - you want qemu_opts_print to be repurposed in
isolation, THEN add the enhancement of a default string.
Everything else in this patch looks okay, though. I'm torn on whether
we need a v27, or whether it is time to just take this as-is, as the end
result of the two patches together is correct, and just the intermediate
state and the split between the patches was awkward.
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
