From: Pablo Neira Ayuso <pablo@netfilter.org>
To: "Carlos Falgueras García" <carlosfg@riseup.net>
Cc: netfilter-devel@vger.kernel.org, kaber@trash.net
Subject: Re: [PATCH 4/4 v4] nftables: rule: Change the field "rule->comment" for an nftnl_udata_buf
Date: Sat, 12 Mar 2016 12:29:55 +0100 [thread overview]
Message-ID: <20160312112955.GC1764@salvia> (raw)
In-Reply-To: <1457645836-10996-4-git-send-email-carlosfg@riseup.net>
On Thu, Mar 10, 2016 at 10:37:16PM +0100, Carlos Falgueras García wrote:
> Now it is possible to store multiple variable length user data into rule.
> Modify the parser in order to fill the nftnl_udata with the comment, and
> the print function for extract these commentary and print it to user.
>
> Signed-off-by: Carlos Falgueras García <carlosfg@riseup.net>
> ---
> include/rule.h | 11 +++++++++--
> src/netlink_delinearize.c | 7 +++++--
> src/netlink_linearize.c | 6 ++++--
> src/parser_bison.y | 14 +++++++++++++-
> src/rule.c | 45 ++++++++++++++++++++++++++++++++++++++++++---
> 5 files changed, 73 insertions(+), 10 deletions(-)
>
> diff --git a/include/rule.h b/include/rule.h
> index c848f0f..c500e88 100644
> --- a/include/rule.h
> +++ b/include/rule.h
> @@ -4,6 +4,7 @@
> #include <stdint.h>
> #include <nftables.h>
> #include <list.h>
> +#include <libnftnl/udata.h>
>
> /**
> * struct handle - handle for tables, chains, rules and sets
> @@ -155,7 +156,7 @@ extern void chain_print_plain(const struct chain *chain);
> * @location: location the rule was defined at
> * @stmt: list of statements
> * @num_stmts: number of statements in stmts list
> - * @comment: comment
> + * @udata: user data
> */
> struct rule {
> struct list_head list;
> @@ -163,7 +164,7 @@ struct rule {
> struct location location;
> struct list_head stmts;
> unsigned int num_stmts;
> - const char *comment;
> + struct nftnl_udata_buf *udata;
I think we should keep the:
const char *comment;
The nftnl_udata_buf should only be used when building the TLV area
from the netlink_linearize step, and when parsing the TLV and covert
it to attribute from netlink_delinearize. Otherwise, we have to deal
with this internal layout from other parts of nft that are not netlink
specific.
> };
>
> extern struct rule *rule_alloc(const struct location *loc,
> @@ -396,4 +397,10 @@ extern int do_command(struct netlink_ctx *ctx, struct cmd *cmd);
> extern int cache_update(enum cmd_ops cmd, struct list_head *msgs);
> extern void cache_release(void);
>
> +enum udata_type {
> + UDATA_TYPE_COMMENT,
> + __UDATA_TYPE_MAX,
> +};
> +#define UDATA_TYPE_MAX (__UDATA_TYPE_MAX - 1)
> +
> #endif /* NFTABLES_RULE_H */
> diff --git a/src/netlink_delinearize.c b/src/netlink_delinearize.c
> index d431588..e655e00 100644
> --- a/src/netlink_delinearize.c
> +++ b/src/netlink_delinearize.c
> @@ -25,6 +25,7 @@
> #include <utils.h>
> #include <erec.h>
> #include <sys/socket.h>
> +#include <libnftnl/udata.h>
>
> struct netlink_parse_ctx {
> struct list_head *msgs;
> @@ -1773,8 +1774,10 @@ struct rule *netlink_delinearize_rule(struct netlink_ctx *ctx,
> uint32_t len;
>
> data = nftnl_rule_get_data(nlr, NFTNL_RULE_USERDATA, &len);
> - pctx->rule->comment = xmalloc(len);
> - memcpy((char *)pctx->rule->comment, data, len);
> + pctx->rule->udata = nftnl_udata_alloc(len);
> + if (!pctx->rule->udata)
> + memory_allocation_error();
> + nftnl_udata_copy_data(pctx->rule->udata, data, len);
Therefore, here you should extract the data from the udata and set
pctx->rule->comment.
> }
>
> nftnl_expr_foreach((struct nftnl_rule *)nlr, netlink_parse_expr, pctx);
> diff --git a/src/netlink_linearize.c b/src/netlink_linearize.c
> index bb51de7..5ac438e 100644
> --- a/src/netlink_linearize.c
> +++ b/src/netlink_linearize.c
> @@ -21,6 +21,7 @@
> #include <netinet/in.h>
>
> #include <linux/netfilter.h>
> +#include <libnftnl/udata.h>
>
>
> struct netlink_linearize_ctx {
> @@ -1171,9 +1172,10 @@ void netlink_linearize_rule(struct netlink_ctx *ctx, struct nftnl_rule *nlr,
> list_for_each_entry(stmt, &rule->stmts, list)
> netlink_gen_stmt(&lctx, stmt);
>
> - if (rule->comment)
> + if (rule->udata)
> nftnl_rule_set_data(nlr, NFTNL_RULE_USERDATA,
> - rule->comment, strlen(rule->comment) + 1);
> + nftnl_udata_data(rule->udata),
> + nftnl_udata_len(rule->udata));
When linearizing, you should do something like:
if (rule->comment) {
udata = nftnl_udata_alloc(NFT_USERDATA_MAXLEN));
if (!udata)
memory_allocation_error();
if (!nftnl_udata_put_strz(udata, UDATA_TYPE_COMMENT, rule->comment))
memory_allocation_error();
nftnl_rule_set_data(nlr, NFTNL_RULE_USERDATA,
nftnl_udata_data(udata),
nftnl_udata_len(udata));
}
>
> netlink_dump_rule(nlr);
> }
> diff --git a/src/parser_bison.y b/src/parser_bison.y
> index 90978ab..9f8c005 100644
> --- a/src/parser_bison.y
> +++ b/src/parser_bison.y
> @@ -24,6 +24,7 @@
> #include <netinet/icmp6.h>
> #include <libnftnl/common.h>
> #include <libnftnl/set.h>
> +#include <libnftnl/udata.h>
>
> #include <rule.h>
> #include <statement.h>
> @@ -1303,7 +1304,18 @@ rule : stmt_list comment_spec
> struct stmt *i;
>
> $$ = rule_alloc(&@$, NULL);
> - $$->comment = $2;
> +
> + if ($2) {
> + if (!($$->udata = nftnl_udata_alloc(NFT_USERDATA_MAXLEN)))
> + memory_allocation_error();
> +
> + if (!nftnl_udata_put_strz($$->udata, UDATA_TYPE_COMMENT, $2)) {
> + erec_queue(error(&@2, "Comment too long: \"%s\"", $2),
> + state->msgs);
> + YYERROR;
> + }
> + }
With the change I propose above, you don't need to update the parser
anymore.
> +
> list_for_each_entry(i, $1, list)
> $$->num_stmts++;
> list_splice_tail($1, &$$->stmts);
> diff --git a/src/rule.c b/src/rule.c
> index 0b78549..112ae85 100644
> --- a/src/rule.c
> +++ b/src/rule.c
> @@ -23,6 +23,7 @@
>
> #include <libnftnl/common.h>
> #include <libnftnl/ruleset.h>
> +#include <libnftnl/udata.h>
> #include <netinet/ip.h>
> #include <linux/netfilter.h>
> #include <linux/netfilter_arp.h>
> @@ -375,21 +376,59 @@ void rule_free(struct rule *rule)
> {
> stmt_list_free(&rule->stmts);
> handle_free(&rule->handle);
> - xfree(rule->comment);
> + nftnl_udata_free(rule->udata);
This change, you will not need it with the change I'm proposing.
> xfree(rule);
> }
>
> +static int rule_parse_udata_cb(const struct nftnl_udata *attr, void *data)
> +{
> + unsigned char *value = nftnl_udata_attr_value(attr);
> + uint8_t type = nftnl_udata_attr_type(attr);
> + uint8_t len = nftnl_udata_attr_len(attr);
> + const struct nftnl_udata **tb = data;
> +
> + switch (type) {
> + case UDATA_TYPE_COMMENT:
> + if (value[len - 1] != '\0')
> + return -1;
> + break;
> + default:
> + break;
> + };
> +
> + tb[type] = attr;
> + return 1;
> +}
> +
> +static void rule_print_udata(const struct nftnl_udata *tb[], uint8_t type)
> +{
> + if (!tb[type])
> + return;
> +
> + switch (type) {
> + case UDATA_TYPE_COMMENT:
> + printf("comment \"%s\" ",
> + (char *)nftnl_udata_attr_value(tb[type]));
> + break;
> + default:
> + break;
> + }
> +}
... And these functions, you will place them in netlink_delinearize.c,
which is where they belong since they are dealing with a low-level
netlink aspect.
> void rule_print(const struct rule *rule)
> {
> const struct stmt *stmt;
> + const struct nftnl_udata *tb[UDATA_TYPE_MAX + 1] = {};
>
> list_for_each_entry(stmt, &rule->stmts, list) {
> stmt->ops->print(stmt);
> printf(" ");
> }
>
> - if (rule->comment)
> - printf("comment \"%s\" ", rule->comment);
> + if (rule->udata) {
> + if (nftnl_udata_parse(rule->udata, rule_parse_udata_cb, tb) > 0)
> + rule_print_udata(tb, UDATA_TYPE_COMMENT);
> + }
With the change I'm proposing above, you will not need this either.
> if (handle_output > 0)
> printf("# handle %" PRIu64, rule->handle.handle);
> --
> 2.7.2
>
> --
> To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2016-03-12 11:30 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-10 21:37 [PATCH 1/4 v4] libnftnl: Implement new buffer of TLV objects Carlos Falgueras García
2016-03-10 21:37 ` [PATCH 2/4 v4] libnftnl: rule: Change the "userdata" attribute to use new TLV buffer Carlos Falgueras García
2016-03-10 21:37 ` [PATCH 3/4 v4] libnftnl: test: Actualize test to check new nftnl_udata features of nftnl_rule Carlos Falgueras García
2016-03-10 21:37 ` [PATCH 4/4 v4] nftables: rule: Change the field "rule->comment" for an nftnl_udata_buf Carlos Falgueras García
2016-03-12 11:29 ` Pablo Neira Ayuso [this message]
2016-03-15 20:29 ` Carlos Falgueras García
2016-03-12 11:09 ` [PATCH 1/4 v4] libnftnl: Implement new buffer of TLV objects Pablo Neira Ayuso
2016-03-15 20:29 ` Carlos Falgueras García
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160312112955.GC1764@salvia \
--to=pablo@netfilter.org \
--cc=carlosfg@riseup.net \
--cc=kaber@trash.net \
--cc=netfilter-devel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).