From: Ido Goshen <Ido@cgstowernetworks.com>
To: Konstantin Ananyev <konstantin.v.ananyev@yandex.ru>,
"dev@dpdk.org" <dev@dpdk.org>
Cc: "stable@dpdk.org" <stable@dpdk.org>
Subject: RE: [PATCH v2 1/2] acl: fix rules with 8 bytes field size are broken
Date: Mon, 16 May 2022 06:23:11 +0000 [thread overview]
Message-ID: <AM0PR09MB397204AF9924EAA2093C4D9BD6CF9@AM0PR09MB3972.eurprd09.prod.outlook.com> (raw)
In-Reply-To: <20220515200319.81002-2-konstantin.v.ananyev@yandex.ru>
Tested-by: Ido Goshen <ido@cgstowernetworks.com>
> -----Original Message-----
> From: Konstantin Ananyev <konstantin.v.ananyev@yandex.ru>
> Sent: Sunday, 15 May 2022 23:03
> To: dev@dpdk.org
> Cc: Ido Goshen <Ido@cgstowernetworks.com>; stable@dpdk.org
> Subject: [PATCH v2 1/2] acl: fix rules with 8 bytes field size are broken
>
> In theory ACL library allows fields with 8B long.
> Though in practice they usually not used, not tested, and as was revealed by Ido,
> this functionality is not working properly.
> There are few places inside ACL build code-path that need to be addressed.
>
> Bugzilla ID: 673
> Fixes: dc276b5780c2 ("acl: new library")
> Cc: stable@dpdk.org
>
> Reported-by: Ido Goshen <ido@cgstowernetworks.com>
> Signed-off-by: Konstantin Ananyev <konstantin.v.ananyev@yandex.ru>
> ---
> lib/acl/acl_bld.c | 14 ++++++++++----
> 1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/lib/acl/acl_bld.c b/lib/acl/acl_bld.c index 7ea30f4186..2816632803
> 100644
> --- a/lib/acl/acl_bld.c
> +++ b/lib/acl/acl_bld.c
> @@ -12,6 +12,9 @@
> /* number of pointers per alloc */
> #define ACL_PTR_ALLOC 32
>
> +/* account for situation when all fields are 8B long */
> +#define ACL_MAX_INDEXES (2 * RTE_ACL_MAX_FIELDS)
> +
> /* macros for dividing rule sets heuristics */
> #define NODE_MAX 0x4000
> #define NODE_MIN 0x800
> @@ -80,7 +83,7 @@ struct acl_build_context {
> struct tb_mem_pool pool;
> struct rte_acl_trie tries[RTE_ACL_MAX_TRIES];
> struct rte_acl_bld_trie bld_tries[RTE_ACL_MAX_TRIES];
> - uint32_t
> data_indexes[RTE_ACL_MAX_TRIES][RTE_ACL_MAX_FIELDS];
> + uint32_t
> data_indexes[RTE_ACL_MAX_TRIES][ACL_MAX_INDEXES];
>
> /* memory free lists for nodes and blocks used for node ptrs */
> struct acl_mem_block blocks[MEM_BLOCK_NUM];
> @@ -988,7 +991,7 @@ build_trie(struct acl_build_context *context, struct
> rte_acl_build_rule *head,
> */
> uint64_t mask;
> mask = RTE_ACL_MASKLEN_TO_BITMASK(
> - fld->mask_range.u32,
> + fld->mask_range.u64,
> rule->config->defs[n].size);
>
> /* gen a mini-trie for this field */ @@ -1301,6
> +1304,9 @@ acl_build_index(const struct rte_acl_config *config, uint32_t
> *data_index)
> if (last_header != config->defs[n].input_index) {
> last_header = config->defs[n].input_index;
> data_index[m++] = config->defs[n].offset;
> + if (config->defs[n].size > sizeof(uint32_t))
> + data_index[m++] = config->defs[n].offset +
> + sizeof(uint32_t);
> }
> }
>
> @@ -1487,7 +1493,7 @@ acl_set_data_indexes(struct rte_acl_ctx *ctx)
> memcpy(ctx->data_indexes + ofs, ctx->trie[i].data_index,
> n * sizeof(ctx->data_indexes[0]));
> ctx->trie[i].data_index = ctx->data_indexes + ofs;
> - ofs += RTE_ACL_MAX_FIELDS;
> + ofs += ACL_MAX_INDEXES;
> }
> }
>
> @@ -1643,7 +1649,7 @@ rte_acl_build(struct rte_acl_ctx *ctx, const struct
> rte_acl_config *cfg)
> /* allocate and fill run-time structures. */
> rc = rte_acl_gen(ctx, bcx.tries, bcx.bld_tries,
> bcx.num_tries, bcx.cfg.num_categories,
> - RTE_ACL_MAX_FIELDS * RTE_DIM(bcx.tries) *
> + ACL_MAX_INDEXES * RTE_DIM(bcx.tries) *
> sizeof(ctx->data_indexes[0]), max_size);
> if (rc == 0) {
> /* set data indexes. */
> --
> 2.34.1
prev parent reply other threads:[~2022-05-16 6:23 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20220426174454.311999-1-konstantin.v.ananyev@yandex.ru>
2022-04-26 17:44 ` [PATCH " Konstantin Ananyev
[not found] ` <20220515200319.81002-1-konstantin.v.ananyev@yandex.ru>
2022-05-15 20:03 ` [PATCH v2 " Konstantin Ananyev
2022-05-16 6:23 ` Ido Goshen [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=AM0PR09MB397204AF9924EAA2093C4D9BD6CF9@AM0PR09MB3972.eurprd09.prod.outlook.com \
--to=ido@cgstowernetworks.com \
--cc=dev@dpdk.org \
--cc=konstantin.v.ananyev@yandex.ru \
--cc=stable@dpdk.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).