Re: [PATCH 1/2] nfs: add support for the umask attribute

[Olga Kornievskaia <aglo@umich.edu>]

On Wed, Feb 1, 2017 at 5:44 PM, Andreas Gruenbacher <agruenba@redhat.com> wrote:
> From: "J. Bruce Fields" <bfields@fieldses.org>
>
> On Wed, Feb 1, 2017 at 10:31 PM, Olga Kornievskaia <aglo@umich.edu> wrote:
>> Any plans to add wireshark support for this?
>
> We did, yes.  Bruce had posted that together with the very first version.  I
> couldn't find the wireshark patch for the current version of the proposal in
> the mailing list archive, so here's that.
>
> Andreas
>
> --
>
> NFSv4.2 umask support
>
> ---
>  epan/dissectors/packet-nfs.c | 19 +++++++++++++++++++
>  1 file changed, 19 insertions(+)
>
> diff --git a/epan/dissectors/packet-nfs.c b/epan/dissectors/packet-nfs.c
> index 6d1dd3b..5f2ce42 100644
> --- a/epan/dissectors/packet-nfs.c
> +++ b/epan/dissectors/packet-nfs.c
> @@ -420,6 +420,7 @@ static int hf_nfs4_fattr_layout_blksize = -1;
>  static int hf_nfs4_fattr_security_label_lfs = -1;
>  static int hf_nfs4_fattr_security_label_pi = -1;
>  static int hf_nfs4_fattr_security_label_context = -1;
> +static int hf_nfs4_fattr_umask_mask = -1;
>  static int hf_nfs4_who = -1;
>  static int hf_nfs4_server = -1;
>  static int hf_nfs4_fslocation = -1;
> @@ -6133,6 +6134,8 @@ static const value_string fattr4_names[] = {
>         {       FATTR4_CHANGE_ATTR_TYPE,   "Change_Attr_Type"           },
>  #define FATTR4_SECURITY_LABEL      80
>         {       FATTR4_SECURITY_LABEL,     "Security_Label"             },
> +#define FATTR4_MODE_UMASK          81
> +       {       FATTR4_MODE_UMASK,         "Mode_Umask"                 },
>         {       0,      NULL    }
>  };
>  static value_string_ext fattr4_names_ext = VALUE_STRING_EXT_INIT(fattr4_names);
> @@ -6717,6 +6720,14 @@ dissect_nfs4_security_label(tvbuff_t *tvb, proto_tree *tree, int offset)
>         return offset;
>  }
>
> +static int
> +dissect_nfs4_mode_umask(tvbuff_t *tvb, proto_tree *tree, int offset)
> +{
> +       offset = dissect_nfs4_mode(tvb, offset, tree);
> +       offset = dissect_rpc_uint32(tvb, tree, hf_nfs4_fattr_umask_mask, offset);
> +       return offset;
> +}
> +
>  #define FATTR4_BITMAP_ONLY 0
>  #define FATTR4_DISSECT_VALUES 1
>
> @@ -7119,6 +7130,10 @@ dissect_nfs4_fattrs(tvbuff_t *tvb, int offset, packet_info *pinfo, proto_tree *t
>                                                 offset = dissect_nfs4_security_label(tvb, attr_tree, offset);
>                                                 break;
>
> +                                       case FATTR4_MODE_UMASK:
> +                                               offset = dissect_nfs4_mode_umask(tvb, attr_tree, offset);
> +                                               break;
> +
>                                         default:
>                                                 break;
>                                         }
> @@ -12509,6 +12524,10 @@ proto_register_nfs(void)
>                         "label_format", "nfs.fattr4.security_label.lfs", FT_UINT32, BASE_DEC,
>                         NULL, 0, NULL, HFILL }},
>
> +               { &hf_nfs4_fattr_umask_mask, {
> +                       "umask", "nfs.fattr4.umask", FT_UINT32, BASE_OCT,
> +                       NULL, 0, NULL, HFILL }},
> +
>                 { &hf_nfs4_fattr_security_label_pi, {
>                         "policy_id", "nfs.fattr4.security_label.pi", FT_UINT32, BASE_DEC,
>                         NULL, 0, NULL, HFILL }},
> --
> 2.7.4


Thank you Andreas. I have tried this patch and it decodes the OPEN
compounds ok. Previously it's been garbage past the unknown attribute
with value 81.