From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Florian Westphal <fw@strlen.de>
Cc: netfilter-devel@vger.kernel.org
Subject: Re: [PATCH nft] json: don't BUG when asked to list synproxies
Date: Thu, 27 Mar 2025 16:49:50 +0100 [thread overview]
Message-ID: <Z-VznsYAZ06V7-EI@calendula> (raw)
In-Reply-To: <20250324115339.11642-1-fw@strlen.de>
On Mon, Mar 24, 2025 at 12:53:36PM +0100, Florian Westphal wrote:
> "-j list synproxys" triggers a BUG().
>
> Rewrite this so that all enum values are handled so the compiler can alert
> us to a missing value in case there are more commands in the future.
>
> While at it, implement a few low-hanging fruites as well.
>
> Not-yet-supported cases are simply ignored.
>
> Signed-off-by: Florian Westphal <fw@strlen.de>
> ---
> src/evaluate.c | 6 ++++--
> src/json.c | 24 ++++++++++++++++++++++--
> src/rule.c | 11 +++++++++--
> 3 files changed, 35 insertions(+), 6 deletions(-)
>
> diff --git a/src/evaluate.c b/src/evaluate.c
> index 785c4fab6b3a..1e7f6f53542b 100644
> --- a/src/evaluate.c
> +++ b/src/evaluate.c
> @@ -6323,7 +6323,9 @@ int cmd_evaluate(struct eval_ctx *ctx, struct cmd *cmd)
> return cmd_evaluate_monitor(ctx, cmd);
> case CMD_IMPORT:
> return cmd_evaluate_import(ctx, cmd);
> - default:
> - BUG("invalid command operation %u\n", cmd->op);
> + case CMD_INVALID:
> + break;
> };
> +
> + BUG("invalid command operation %u\n", cmd->op);
> }
> diff --git a/src/json.c b/src/json.c
> index 96413d70895a..f92f86bbc974 100644
> --- a/src/json.c
> +++ b/src/json.c
> @@ -1971,7 +1971,7 @@ static json_t *generate_json_metainfo(void)
> int do_command_list_json(struct netlink_ctx *ctx, struct cmd *cmd)
> {
> struct table *table = NULL;
> - json_t *root;
> + json_t *root = NULL;
>
> if (cmd->handle.table.name)
> table = table_cache_find(&ctx->nft->cache.table_cache,
> @@ -2026,6 +2026,13 @@ int do_command_list_json(struct netlink_ctx *ctx, struct cmd *cmd)
> case CMD_OBJ_CT_HELPERS:
> root = do_list_obj_json(ctx, cmd, NFT_OBJECT_CT_HELPER);
> break;
> + case CMD_OBJ_CT_TIMEOUT:
> + case CMD_OBJ_CT_TIMEOUTS:
> + root = do_list_obj_json(ctx, cmd, NFT_OBJECT_CT_TIMEOUT);
> + case CMD_OBJ_CT_EXPECT:
> + case CMD_OBJ_CT_EXPECTATIONS:
> + root = do_list_obj_json(ctx, cmd, NFT_OBJECT_CT_EXPECT);
> + break;
> case CMD_OBJ_LIMIT:
> case CMD_OBJ_LIMITS:
> root = do_list_obj_json(ctx, cmd, NFT_OBJECT_LIMIT);
> @@ -2034,14 +2041,27 @@ int do_command_list_json(struct netlink_ctx *ctx, struct cmd *cmd)
> case CMD_OBJ_SECMARKS:
> root = do_list_obj_json(ctx, cmd, NFT_OBJECT_SECMARK);
> break;
> + case CMD_OBJ_SYNPROXY:
> + case CMD_OBJ_SYNPROXYS:
> + root = do_list_obj_json(ctx, cmd, NFT_OBJECT_SYNPROXY);
> + break;
> case CMD_OBJ_FLOWTABLE:
> root = do_list_flowtable_json(ctx, cmd, table);
> break;
> case CMD_OBJ_FLOWTABLES:
> root = do_list_flowtables_json(ctx, cmd);
> break;
> - default:
> + case CMD_OBJ_HOOKS:
> + case CMD_OBJ_MONITOR:
> + case CMD_OBJ_MARKUP:
> + case CMD_OBJ_SETELEMS:
> + case CMD_OBJ_RULE:
> + case CMD_OBJ_EXPR:
> + case CMD_OBJ_ELEMENTS:
> + return 0;
> + case CMD_OBJ_INVALID:
> BUG("invalid command object type %u\n", cmd->obj);
> + break;
> }
>
> if (!json_is_array(root)) {
> diff --git a/src/rule.c b/src/rule.c
> index 00fbbc4c080a..cf895a5acf5b 100644
> --- a/src/rule.c
> +++ b/src/rule.c
> @@ -2445,10 +2445,17 @@ static int do_command_list(struct netlink_ctx *ctx, struct cmd *cmd)
> return do_list_flowtables(ctx, cmd);
> case CMD_OBJ_HOOKS:
> return do_list_hooks(ctx, cmd);
> - default:
> - BUG("invalid command object type %u\n", cmd->obj);
> + case CMD_OBJ_MONITOR:
monitor with list make no sense, maybe report error here instead.
> + case CMD_OBJ_MARKUP:
CMD_OBJ_MARKUP is obsolete / early xml support that was removed, this
has been deprecated for long time, I can follow up to remove it in a
follow up patch, there is just a few remaining scaffolding around
this.
> + case CMD_OBJ_SETELEMS:
this is an internal command, it should not ever be seen in this path.
> + case CMD_OBJ_EXPR:
This is for the describe command, it can never happen with list.
> + case CMD_OBJ_ELEMENTS:
support for add, get, delete and destroy.
... In summary, I would return here:
errno = EOPNOTSUPP;
return -1;
Agreed?
> + return 0;
> + case CMD_OBJ_INVALID:
> + break;
> }
>
> + BUG("invalid command object type %u\n", cmd->obj);
> return 0;
> }
>
> --
> 2.48.1
>
>
prev parent reply other threads:[~2025-03-27 15:49 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-24 11:53 [PATCH nft] json: don't BUG when asked to list synproxies Florian Westphal
2025-03-27 15:49 ` Pablo Neira Ayuso [this message]
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=Z-VznsYAZ06V7-EI@calendula \
--to=pablo@netfilter.org \
--cc=fw@strlen.de \
--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).