| Message ID | 20210302171623.49709-1-kwolf@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | qom: Check for wellformed id in user_creatable_add_type() | expand |
On 3/2/21 11:16 AM, Kevin Wolf wrote: > Most code paths for creating a user creatable object go through > QemuOpts, which ensures that the provided 'id' option is actually a > valid identifier. > > However, there are some code paths that don't go through QemuOpts: > qemu-storage-daemon --object (since commit 8db1efd3) and QMP object-add > (since it was first introduced in commit cff8b2c6). We need to have the > same validity check for those, too. > > This adds the check and makes it print the same error message as > QemuOpts on failure. > > Signed-off-by: Kevin Wolf <kwolf@redhat.com> > --- > This makes sense even without the -object QAPIfication, so no reason to > wait for v3 of that series to get this fixed. > Reviewed-by: Eric Blake <eblake@redhat.com>
On 02/03/21 18:16, Kevin Wolf wrote: > Most code paths for creating a user creatable object go through > QemuOpts, which ensures that the provided 'id' option is actually a > valid identifier. > > However, there are some code paths that don't go through QemuOpts: > qemu-storage-daemon --object (since commit 8db1efd3) and QMP object-add > (since it was first introduced in commit cff8b2c6). We need to have the > same validity check for those, too. > > This adds the check and makes it print the same error message as > QemuOpts on failure. > > Signed-off-by: Kevin Wolf <kwolf@redhat.com> > --- > This makes sense even without the -object QAPIfication, so no reason to > wait for v3 of that series to get this fixed. > > qom/object_interfaces.c | 9 +++++++++ > 1 file changed, 9 insertions(+) > > diff --git a/qom/object_interfaces.c b/qom/object_interfaces.c > index 1e9ad6f08a..515ca4557e 100644 > --- a/qom/object_interfaces.c > +++ b/qom/object_interfaces.c > @@ -8,6 +8,7 @@ > #include "qapi/qobject-input-visitor.h" > #include "qom/object_interfaces.h" > #include "qemu/help_option.h" > +#include "qemu/id.h" > #include "qemu/module.h" > #include "qemu/option.h" > #include "qapi/opts-visitor.h" > @@ -41,11 +42,19 @@ Object *user_creatable_add_type(const char *type, const char *id, > const QDict *qdict, > Visitor *v, Error **errp) > { > + ERRP_GUARD(); > Object *obj; > ObjectClass *klass; > const QDictEntry *e; > Error *local_err = NULL; > > + if (!id_wellformed(id)) { > + error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "id", "an identifier"); > + error_append_hint(errp, "Identifiers consist of letters, digits, " > + "'-', '.', '_', starting with a letter.\n"); > + return NULL; > + } > + > klass = object_class_by_name(type); > if (!klass) { > error_setg(errp, "invalid object type: %s", type); > Queued, thanks. Paolo
On 02/03/21 18:16, Kevin Wolf wrote: > Most code paths for creating a user creatable object go through > QemuOpts, which ensures that the provided 'id' option is actually a > valid identifier. > > However, there are some code paths that don't go through QemuOpts: > qemu-storage-daemon --object (since commit 8db1efd3) and QMP object-add > (since it was first introduced in commit cff8b2c6). We need to have the > same validity check for those, too. > > This adds the check and makes it print the same error message as > QemuOpts on failure. > > Signed-off-by: Kevin Wolf <kwolf@redhat.com> > --- > This makes sense even without the -object QAPIfication, so no reason to > wait for v3 of that series to get this fixed. It needs a check for id != NULL, but no big deal so I added it. Thanks, Paolo > qom/object_interfaces.c | 9 +++++++++ > 1 file changed, 9 insertions(+) > > diff --git a/qom/object_interfaces.c b/qom/object_interfaces.c > index 1e9ad6f08a..515ca4557e 100644 > --- a/qom/object_interfaces.c > +++ b/qom/object_interfaces.c > @@ -8,6 +8,7 @@ > #include "qapi/qobject-input-visitor.h" > #include "qom/object_interfaces.h" > #include "qemu/help_option.h" > +#include "qemu/id.h" > #include "qemu/module.h" > #include "qemu/option.h" > #include "qapi/opts-visitor.h" > @@ -41,11 +42,19 @@ Object *user_creatable_add_type(const char *type, const char *id, > const QDict *qdict, > Visitor *v, Error **errp) > { > + ERRP_GUARD(); > Object *obj; > ObjectClass *klass; > const QDictEntry *e; > Error *local_err = NULL; > > + if (!id_wellformed(id)) { > + error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "id", "an identifier"); > + error_append_hint(errp, "Identifiers consist of letters, digits, " > + "'-', '.', '_', starting with a letter.\n"); > + return NULL; > + } > + > klass = object_class_by_name(type); > if (!klass) { > error_setg(errp, "invalid object type: %s", type); >
Am 06.03.2021 um 11:50 hat Paolo Bonzini geschrieben: > On 02/03/21 18:16, Kevin Wolf wrote: > > Most code paths for creating a user creatable object go through > > QemuOpts, which ensures that the provided 'id' option is actually a > > valid identifier. > > > > However, there are some code paths that don't go through QemuOpts: > > qemu-storage-daemon --object (since commit 8db1efd3) and QMP object-add > > (since it was first introduced in commit cff8b2c6). We need to have the > > same validity check for those, too. > > > > This adds the check and makes it print the same error message as > > QemuOpts on failure. > > > > Signed-off-by: Kevin Wolf <kwolf@redhat.com> > > --- > > This makes sense even without the -object QAPIfication, so no reason to > > wait for v3 of that series to get this fixed. > > It needs a check for id != NULL, but no big deal so I added it. Oops, yes. Thanks for fixing it up. Kevin
diff --git a/qom/object_interfaces.c b/qom/object_interfaces.c index 1e9ad6f08a..515ca4557e 100644 --- a/qom/object_interfaces.c +++ b/qom/object_interfaces.c @@ -8,6 +8,7 @@ #include "qapi/qobject-input-visitor.h" #include "qom/object_interfaces.h" #include "qemu/help_option.h" +#include "qemu/id.h" #include "qemu/module.h" #include "qemu/option.h" #include "qapi/opts-visitor.h" @@ -41,11 +42,19 @@ Object *user_creatable_add_type(const char *type, const char *id, const QDict *qdict, Visitor *v, Error **errp) { + ERRP_GUARD(); Object *obj; ObjectClass *klass; const QDictEntry *e; Error *local_err = NULL; + if (!id_wellformed(id)) { + error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "id", "an identifier"); + error_append_hint(errp, "Identifiers consist of letters, digits, " + "'-', '.', '_', starting with a letter.\n"); + return NULL; + } + klass = object_class_by_name(type); if (!klass) { error_setg(errp, "invalid object type: %s", type);
Most code paths for creating a user creatable object go through QemuOpts, which ensures that the provided 'id' option is actually a valid identifier. However, there are some code paths that don't go through QemuOpts: qemu-storage-daemon --object (since commit 8db1efd3) and QMP object-add (since it was first introduced in commit cff8b2c6). We need to have the same validity check for those, too. This adds the check and makes it print the same error message as QemuOpts on failure. Signed-off-by: Kevin Wolf <kwolf@redhat.com> --- This makes sense even without the -object QAPIfication, so no reason to wait for v3 of that series to get this fixed. qom/object_interfaces.c | 9 +++++++++ 1 file changed, 9 insertions(+)