* Re: PATCH: add IP_USER_FLOW to rxntuple interface
[not found] <OF20F9D953.0F90C5C5-ON85257885.004A9AEF-85257885.004A9AF3@BeldenCDT.com>
@ 2011-05-03 15:59 ` Alexander Duyck
0 siblings, 0 replies; only message in thread
From: Alexander Duyck @ 2011-05-03 15:59 UTC (permalink / raw)
To: Sebastian.Poehn@Belden.com; +Cc: netdev@vger.kernel.org, Ben Hutchings
On 5/3/2011 6:34 AM, Sebastian.Poehn@Belden.com wrote:
> I added IP_USER_FLOW to the rx ntuple interface of ethtool which will
> support raw IP traffic classification. If the patch is okay i will
> add the new parameters to the manpage.
>
> Signed-off-by: Sebastian Poehn<sebastian.poehn@belden.com>
>
> --- ethtool/ethtool.c.orig 2011-05-03 11:17:30.457012539 +0200
> +++ ethtool/ethtool.c 2011-05-03 13:24:52.709010424 +0200
> @@ -396,6 +396,12 @@ static int ntuple_psrc_seen = 0;
> static int ntuple_psrc_mask_seen = 0;
> static int ntuple_pdst_seen = 0;
> static int ntuple_pdst_mask_seen = 0;
> +static int ntuple_l4_4_bytes_seen = 0;
> +static int ntuple_l4_4_bytes_mask_seen = 0;
> +static int ntuple_ip_ver_seen = 0;
> +static int ntuple_ip_ver_mask_seen = 0;
> +static int ntuple_ip_proto_seen = 0;
> +static int ntuple_ip_proto_mask_seen = 0;
> static int ntuple_ether_dst_seen = 0;
> static int ntuple_ether_dst_mask_seen = 0;
> static int ntuple_ether_src_seen = 0;
> @@ -423,10 +429,12 @@ typedef enum {
> CMDL_NONE,
> CMDL_BOOL,
> CMDL_S32,
> + CMDL_U8,
> CMDL_U16,
> CMDL_U32,
> CMDL_U64,
> CMDL_BE16,
> + CMDL_BE32,
> CMDL_IP4,
> CMDL_STR,
> CMDL_FLAG,
> @@ -554,6 +562,38 @@ static struct cmdline_info cmdline_ntupl
> { "action", CMDL_S32,&ntuple_fs.action, NULL },
> };
>
> +static struct cmdline_info cmdline_ntuple_usr_ip4[] = {
> + { "src-ip", CMDL_IP4,&ntuple_fs.h_u.usr_ip4_spec.ip4src, NULL,
> + 0,&ntuple_ip4src_seen },
> + { "src-ip-mask", CMDL_IP4,&ntuple_fs.m_u.usr_ip4_spec.ip4src, NULL,
> + 0,&ntuple_ip4src_mask_seen },
> + { "dst-ip", CMDL_IP4,&ntuple_fs.h_u.usr_ip4_spec.ip4dst, NULL,
> + 0,&ntuple_ip4dst_seen },
> + { "dst-ip-mask", CMDL_IP4,&ntuple_fs.m_u.usr_ip4_spec.ip4dst, NULL,
> + 0,&ntuple_ip4dst_mask_seen },
> + { "l4-data", CMDL_BE32,&ntuple_fs.h_u.usr_ip4_spec.l4_4_bytes, NULL,
> + 0,&ntuple_l4_4_bytes_seen },
> + { "l4-data-mask", CMDL_BE32,&ntuple_fs.m_u.usr_ip4_spec.l4_4_bytes, NULL,
> + 0,&ntuple_l4_4_bytes_mask_seen },
> + { "version", CMDL_U8,&ntuple_fs.h_u.usr_ip4_spec.ip_ver, NULL,
> + 0,&ntuple_ip_ver_seen },
> + { "version-mask", CMDL_U8,&ntuple_fs.m_u.usr_ip4_spec.ip_ver, NULL,
> + 0,&ntuple_ip_ver_mask_seen },
> + { "iana", CMDL_U8,&ntuple_fs.h_u.usr_ip4_spec.proto, NULL,
> + 0,&ntuple_ip_proto_seen },
> + { "iana-mask", CMDL_U8,&ntuple_fs.m_u.usr_ip4_spec.proto, NULL,
> + 0,&ntuple_ip_proto_mask_seen },
> + { "vlan", CMDL_U16,&ntuple_fs.vlan_tag, NULL,
> + 0,&ntuple_vlan_tag_seen },
> + { "vlan-mask", CMDL_U16,&ntuple_fs.vlan_tag_mask, NULL,
> + 0,&ntuple_vlan_tag_mask_seen },
> + { "user-def", CMDL_U64,&ntuple_fs.data, NULL,
> + 0,&ntuple_user_def_seen },
> + { "user-def-mask", CMDL_U64,&ntuple_fs.data_mask, NULL,
> + 0,&ntuple_user_def_mask_seen },
> + { "action", CMDL_S32,&ntuple_fs.action, NULL },
> +};
> +
> static struct cmdline_info cmdline_ntuple_ether[] = {
> { "dst", CMDL_MAC, ntuple_fs.h_u.ether_spec.h_dest, NULL,
> 0,&ntuple_ether_dst_seen },
> @@ -689,6 +729,11 @@ static void parse_generic_cmdline(int ar
> 0x7fffffff);
> break;
> }
> + case CMDL_U8: {
> + u8 *p = info[idx].wanted_val;
> + *p = get_uint_range(argp[i], 0, 0xff);
> + break;
> + }
> case CMDL_U16: {
> u16 *p = info[idx].wanted_val;
> *p = get_uint_range(argp[i], 0, 0xffff);
> @@ -714,6 +759,13 @@ static void parse_generic_cmdline(int ar
> 0xffff));
> break;
> }
> + case CMDL_BE32: {
> + u32 *p = info[idx].wanted_val;
> + *p = cpu_to_be32(
> + get_uint_range(argp[i], 0,
> + 0xffffffff));
> + break;
> + }
> case CMDL_IP4: {
> u32 *p = info[idx].wanted_val;
> struct in_addr in;
> @@ -786,6 +838,8 @@ static int rxflow_str_to_type(const char
> flow_type = AH_ESP_V4_FLOW;
> else if (!strcmp(str, "sctp4"))
> flow_type = SCTP_V4_FLOW;
> + else if (!strcmp(str, "usr4"))
> + flow_type = IP_USER_FLOW;
> else if (!strcmp(str, "tcp6"))
> flow_type = TCP_V6_FLOW;
> else if (!strcmp(str, "udp6"))
> @@ -1648,6 +1702,23 @@ static void parse_rxntupleopts(int argc,
> if (!ntuple_ether_proto_seen)
> ntuple_fs.m_u.ether_spec.h_proto = 0xffff;
> break;
> + case IP_USER_FLOW:
> + parse_generic_cmdline(argc, argp, i + 1,
> +&sntuple_changed,
> + cmdline_ntuple_usr_ip4,
> + ARRAY_SIZE(cmdline_ntuple_usr_ip4));
> + if (!ntuple_ip4src_seen)
> + ntuple_fs.m_u.usr_ip4_spec.ip4src = 0xffffffff;
> + if (!ntuple_ip4dst_seen)
> + ntuple_fs.m_u.usr_ip4_spec.ip4dst = 0xffffffff;
> + if (!ntuple_l4_4_bytes_seen)
> + ntuple_fs.m_u.usr_ip4_spec.l4_4_bytes = 0xffffffff;
> + if (!ntuple_ip_ver_seen)
> + ntuple_fs.m_u.usr_ip4_spec.ip_ver = 0xff;
> + if (!ntuple_ip_proto_seen)
> + ntuple_fs.m_u.usr_ip4_spec.proto = 0xff;
> + ntuple_fs.m_u.tcp_ip4_spec.tos = 0xff;
> + break;
> default:
> fprintf(stderr, "Unsupported flow type \"%s\"\n", argp[i]);
> exit(106);
>
All of this code doesn't exist after the changes I am making to add
network flow classifier support to ethtool. I also believe this is all
redundant since the updated changes should allow ntuple to use
IP_USER_FLOW since it already supported in network flow classifier.
Thanks,
Alex
^ permalink raw reply [flat|nested] only message in thread