Re: [Qemu-devel] [PATCH 1/3] i386: fix regression parsing multiboot initrd modules

["Daniel P. Berrangé" <berrange@redhat.com>]

On Fri, May 18, 2018 at 06:54:48PM +0100, Peter Maydell wrote:
> On 14 May 2018 at 18:19, Daniel P. Berrangé <berrange@redhat.com> wrote:
> > The logic for parsing the multiboot initrd modules was messed up in
> >
> >   commit 950c4e6c94b15cd0d8b63891dddd7a8dbf458e6a
> >   Author: Daniel P. Berrangé <berrange@redhat.com>
> >   Date:   Mon Apr 16 12:17:43 2018 +0100
> >
> >     opts: don't silently truncate long option values
> >
> > Causing the length to be undercounter, and the number of modules over
> > counted. It also passes NULL to get_opt_value() which was not robust
> > at accepting a NULL value.
> >
> > Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
> > ---
> >  hw/i386/multiboot.c | 3 +--
> >  util/qemu-option.c  | 4 +++-
> >  2 files changed, 4 insertions(+), 3 deletions(-)
> >
> > diff --git a/hw/i386/multiboot.c b/hw/i386/multiboot.c
> > index 7a2953e26f..8e26545814 100644
> > --- a/hw/i386/multiboot.c
> > +++ b/hw/i386/multiboot.c
> > @@ -292,8 +292,7 @@ int load_multiboot(FWCfgState *fw_cfg,
> >      cmdline_len += strlen(kernel_cmdline) + 1;
> >      if (initrd_filename) {
> >          const char *r = get_opt_value(initrd_filename, NULL);
> > -        cmdline_len += strlen(r) + 1;
> > -        mbs.mb_mods_avail = 1;
> > +        cmdline_len += strlen(initrd_filename) + 1;
> >          while (1) {
> >              mbs.mb_mods_avail++;
> >              r = get_opt_value(r, NULL);
> > diff --git a/util/qemu-option.c b/util/qemu-option.c
> > index 58d1c23893..8a68bc2314 100644
> > --- a/util/qemu-option.c
> > +++ b/util/qemu-option.c
> > @@ -75,7 +75,9 @@ const char *get_opt_value(const char *p, char **value)
> >      size_t capacity = 0, length;
> >      const char *offset;
> >
> > -    *value = NULL;
> > +    if (value) {
> > +        *value = NULL;
> > +    }
> >      while (1) {
> >          offset = strchr(p, ',');
> >          if (!offset) {
> 
> Don't we delete this check again in patch 3? If we're
> going to fix this by making multiboot.c not pass in NULL pointers,
> is there a reason not to simply do that?

That is correct, however, I wanted to do a really simple fix that addressed
the crasher regression, so that aspect of the fix wasn't obscured by the
bigger refactoring I do in 2 & 3.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|