From: David Marchand <david.marchand@redhat.com>
To: Chengwen Feng <fengchengwen@huawei.com>
Cc: thomas@monjalon.net, dev@dpdk.org
Subject: Re: [PATCH 3/4] argparse: fix argument flags operate as uint32 type
Date: Wed, 6 Mar 2024 18:34:11 +0100 [thread overview]
Message-ID: <CAJFAV8xw3ie7omZNQ4aHAidyjnMhVQWpZrZpZQeN9wOTCd9YMw@mail.gmail.com> (raw)
In-Reply-To: <20240220131502.47510-4-fengchengwen@huawei.com>
On Tue, Feb 20, 2024 at 2:16 PM Chengwen Feng <fengchengwen@huawei.com> wrote:
>
> The struct rte_argparse_arg's flags was 64bit type, uint64_t should be
> used instead of uint32_t where the operation happened.
Something is strange.
An enum in C is represented as an int.
Plus, this enum type is not used anywhere:
lib/argparse/rte_argparse.h:enum rte_argparse_flag {
lib/argparse/rte_argparse.h: /** @see rte_argparse_flag */
I understand the flags are a bitmask.
So please remove this enum and define macros instead.
>
> Also, the flags' bit16 was also unused, so don't test bit16 in testcase
> test_argparse_invalid_arg_flags.
>
> Fixes: 6c5c6571601c ("argparse: verify argument config")
> Fixes: 31ed9f9f43bb ("argparse: parse parameters")
>
> Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
> ---
> app/test/test_argparse.c | 16 ++++++++--------
> lib/argparse/rte_argparse.c | 4 ++--
> 2 files changed, 10 insertions(+), 10 deletions(-)
>
> diff --git a/app/test/test_argparse.c b/app/test/test_argparse.c
> index c98bcee56d..708a575e16 100644
> --- a/app/test/test_argparse.c
> +++ b/app/test/test_argparse.c
> @@ -188,7 +188,7 @@ test_argparse_invalid_arg_help(void)
> static int
> test_argparse_invalid_has_val(void)
> {
> - uint32_t set_mask[] = { 0,
> + uint64_t set_mask[] = { 0,
> RTE_ARGPARSE_ARG_NO_VALUE,
> RTE_ARGPARSE_ARG_OPTIONAL_VALUE
> };
> @@ -197,7 +197,7 @@ test_argparse_invalid_has_val(void)
> int ret;
>
> obj = test_argparse_init_obj();
> - obj->args[0].flags &= ~0x3u;
> + obj->args[0].flags &= ~0x3ull;
If flags is a uint64_t, use RTE_BIT64().
I don't know the argparse API, but why do we need this hardcoded (and
hard to understand) ~3 value?
Can it be expressed with the flags defined in the API?
> ret = rte_argparse_parse(obj, default_argc, default_argv);
> TEST_ASSERT(ret == -EINVAL, "Argparse parse expect failed!");
>
> @@ -205,7 +205,7 @@ test_argparse_invalid_has_val(void)
> obj = test_argparse_init_obj();
> obj->args[0].name_long = "abc";
> obj->args[0].name_short = NULL;
> - obj->args[0].flags &= ~0x3u;
> + obj->args[0].flags &= ~0x3ull;
> obj->args[0].flags |= set_mask[index];
> ret = rte_argparse_parse(obj, default_argc, default_argv);
> TEST_ASSERT(ret == -EINVAL, "Argparse parse expect failed!");
> @@ -269,7 +269,7 @@ test_argparse_invalid_arg_flags(void)
> int ret;
>
> obj = test_argparse_init_obj();
> - obj->args[0].flags |= ~0x107FFu;
> + obj->args[0].flags |= ~0x7FFull;
Same comments as above.
--
David Marchand
next prev parent reply other threads:[~2024-03-06 17:34 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-20 13:14 [PATCH 0/4] refine argparse library Chengwen Feng
2024-02-20 13:14 ` [PATCH 1/4] argparse: refine error message Chengwen Feng
2024-02-20 13:15 ` [PATCH 2/4] argparse: remove dead code Chengwen Feng
2024-02-20 13:15 ` [PATCH 3/4] argparse: fix argument flags operate as uint32 type Chengwen Feng
2024-03-06 17:34 ` David Marchand [this message]
2024-03-07 13:14 ` fengchengwen
2024-02-20 13:15 ` [PATCH 4/4] test/argparse: refine testcases Chengwen Feng
2024-03-07 13:07 ` [PATCH v2 0/5] refine argparse library Chengwen Feng
2024-03-07 13:07 ` [PATCH v2 1/5] argparse: refine error message Chengwen Feng
2024-03-07 13:07 ` [PATCH v2 2/5] argparse: remove dead code Chengwen Feng
2024-03-07 13:07 ` [PATCH v2 3/5] argparse: replace flag enum with marco Chengwen Feng
2024-03-07 17:43 ` Tyler Retzlaff
2024-03-07 17:58 ` David Marchand
2024-03-07 18:37 ` Tyler Retzlaff
2024-03-07 13:07 ` [PATCH v2 4/5] argparse: fix argument flags operate as uint32 type Chengwen Feng
2024-03-18 2:31 ` Thomas Monjalon
2024-03-18 9:22 ` fengchengwen
2024-03-07 13:07 ` [PATCH v2 5/5] test/argparse: refine testcases Chengwen Feng
2024-03-18 9:18 ` [PATCH v3 0/6] refine argparse library Chengwen Feng
2024-03-18 9:18 ` [PATCH v3 1/6] argparse: refine error message Chengwen Feng
2024-03-18 9:18 ` [PATCH v3 2/6] argparse: remove dead code Chengwen Feng
2024-03-18 9:18 ` [PATCH v3 3/6] argparse: replace flag enum with marco Chengwen Feng
2024-03-18 9:18 ` [PATCH v3 4/6] argparse: fix argument flags operate as uint32 type Chengwen Feng
2024-03-18 9:38 ` Thomas Monjalon
2024-03-18 11:24 ` fengchengwen
2024-03-18 9:18 ` [PATCH v3 5/6] test/argparse: refine testcases Chengwen Feng
2024-03-18 9:18 ` [PATCH v3 6/6] argparse: fix doc don't display two hyphens Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 0/6] refine argparse library Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 1/6] argparse: refine error message Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 2/6] argparse: remove dead code Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 3/6] argparse: replace flag enum with marco Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 4/6] argparse: fix argument flags operate as uint32 type Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 5/6] test/argparse: refine testcases Chengwen Feng
2024-03-18 11:18 ` [PATCH v4 6/6] argparse: fix doc don't display two hyphens Chengwen Feng
2024-03-21 17:27 ` [PATCH v4 0/6] refine argparse library Thomas Monjalon
2024-03-22 1:24 ` fengchengwen
2024-05-08 9:10 ` fengchengwen
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=CAJFAV8xw3ie7omZNQ4aHAidyjnMhVQWpZrZpZQeN9wOTCd9YMw@mail.gmail.com \
--to=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=fengchengwen@huawei.com \
--cc=thomas@monjalon.net \
/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).