From: huangdengdui <huangdengdui@huawei.com>
To: "Morten Brørup" <mb@smartsharesystems.com>, dev@dpdk.org
Cc: <stephen@networkplumber.org>, <david.marchand@redhat.com>,
<roretzla@linux.microsoft.com>, <lihuisong@huawei.com>,
<fengchengwen@huawei.com>, <liuyonglong@huawei.com>
Subject: Re: [PATCH v4] eal: fix uncheck lcore ID and lcore role
Date: Thu, 7 Aug 2025 10:43:09 +0800 [thread overview]
Message-ID: <079a2899-b2c4-4e50-b67a-eaff5b04e9ec@huawei.com> (raw)
In-Reply-To: <98CBD80474FA8B44BF855DF32C47DC35E9FDC0@smartserver.smartshare.dk>
On 2025/7/24 22:44, Morten Brørup wrote:
>> From: Dengdui Huang [mailto:huangdengdui@huawei.com]
>> Sent: Thursday, 24 July 2025 13.26
>>
>> The worker_id may come from user input. So it is necessary to verify it.
>>
>> In addition, only the lcore whose role is ROLE_RTE or ROLE_SERVICE can
>> be
>> launched. This patch adds the lcore role check.
>>
>> Fixes: a95d70547c57 ("eal: factorize lcore main loop")
>> Cc: stable@dpdk.org
>>
>> Signed-off-by: Dengdui Huang <huangdengdui@huawei.com>
>> ---
>> lib/eal/common/eal_common_launch.c | 18 ++++++++++++++++++
>> 1 file changed, 18 insertions(+)
>>
>> diff --git a/lib/eal/common/eal_common_launch.c
>> b/lib/eal/common/eal_common_launch.c
>> index a7deac6ecd..09366ea404 100644
>> --- a/lib/eal/common/eal_common_launch.c
>> +++ b/lib/eal/common/eal_common_launch.c
>> @@ -20,6 +20,9 @@ RTE_EXPORT_SYMBOL(rte_eal_wait_lcore)
>> int
>> rte_eal_wait_lcore(unsigned worker_id)
>> {
>> + if (unlikely(worker_id >= RTE_MAX_LCORE))
>> + return -EINVAL;
>> +
>> while (rte_atomic_load_explicit(&lcore_config[worker_id].state,
>> rte_memory_order_acquire) != WAIT)
>> rte_pause();
>> @@ -37,6 +40,18 @@ int
>> rte_eal_remote_launch(lcore_function_t *f, void *arg, unsigned int
>> worker_id)
>> {
>> int rc = -EBUSY;
>> + uint8_t role;
>
> Use the correct type:
> enum rte_lcore_role_t role;
OK, I made a mistake, and I will fix it in the next version.
>
>> +
>> + if (unlikely(worker_id >= RTE_MAX_LCORE)) {
>> + rc = -EINVAL;
>> + goto finish;
>> + }
>> +
>> + role = lcore_config[worker_id].core_role;
>> + if (role != ROLE_RTE && role != ROLE_SERVICE) {
>> + rc = -EINVAL;
>> + goto finish;
>> + }
>>
>> /* Check if the worker is in 'WAIT' state. Use acquire order
>> * since 'state' variable is used as the guard variable.
>> @@ -98,6 +113,9 @@ RTE_EXPORT_SYMBOL(rte_eal_get_lcore_state)
>> enum rte_lcore_state_t
>> rte_eal_get_lcore_state(unsigned lcore_id)
>> {
>> + if (unlikely(lcore_id >= RTE_MAX_LCORE))
>> + return -EINVAL;
>
> The return type of this function is enum rte_lcore_state_t, so it cannot return (int)-EINVAL.
>
> Maybe it's better to update the function's documentation instead of its implementation.
> Like the documentation of the lcore_id parameter passed to rte_eal_lcore_role():
> https://elixir.bootlin.com/dpdk/v25.07/source/lib/eal/include/rte_lcore.h#L44
>
> After further consideration, updating the documentation might also be a better solution for the other functions.
This is a better solution, and I will do it in the next version.
>
> If you add a new return value (-EINVAL) to a function, it must be listed in function's documentation, and in the documentation of functions calling that function.
>
>> +
>> return lcore_config[lcore_id].state;
>> }
>>
>> --
>> 2.33.0
>
>
next prev parent reply other threads:[~2025-08-07 2:43 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-21 4:03 [PATCH 0/2] fix the problem of dma-perf infinite loop Dengdui Huang
2025-03-21 4:03 ` [PATCH 1/2] eal: fix wake a incorrect lcore Dengdui Huang
2025-03-21 4:03 ` [PATCH 2/2] app/dma-perf: fix infinite loop Dengdui Huang
2025-03-21 15:58 ` Stephen Hemminger
2025-03-24 3:42 ` huangdengdui
2025-03-21 7:49 ` [PATCH 0/2] fix the problem of dma-perf " David Marchand
2025-03-24 3:42 ` huangdengdui
2025-03-27 8:58 ` [PATCH] app/dma-perf: fix " Dengdui Huang
2025-03-27 8:58 ` [PATCH 1/2] eal: fix uncheck worker ID Dengdui Huang
2025-03-27 8:58 ` [PATCH 2/2] app/dma-perf: fix infinite loop Dengdui Huang
2025-03-27 9:01 ` [PATCH v2 0/2] fix the problem of dma-perf " Dengdui Huang
2025-03-27 9:01 ` [PATCH v2 1/2] eal: fix uncheck worker ID Dengdui Huang
2025-03-27 9:32 ` Morten Brørup
2025-04-02 8:28 ` huangdengdui
2025-03-27 9:01 ` [PATCH v2 2/2] app/dma-perf: fix infinite loop Dengdui Huang
2025-04-02 12:24 ` [PATCH v3 0/3] fix the problem of dma-perf " Dengdui Huang
2025-04-02 12:24 ` [PATCH v3 1/3] eal: fix uncheck lcore ID Dengdui Huang
2025-04-02 12:24 ` [PATCH v3 2/3] eal: fix unckeck pipe Dengdui Huang
2025-04-02 14:45 ` Morten Brørup
2025-04-03 8:12 ` huangdengdui
2025-04-02 12:24 ` [PATCH v3 3/3] app/dma-perf: fix infinite loop Dengdui Huang
2025-07-24 11:25 ` [PATCH v4] eal: fix uncheck lcore ID and lcore role Dengdui Huang
2025-07-24 14:44 ` Morten Brørup
2025-08-07 2:43 ` huangdengdui [this message]
2025-08-07 2:49 ` [PATCH v5] eal: fix uncheck " Dengdui Huang
2025-08-07 3:15 ` Stephen Hemminger
2025-08-07 3:57 ` huangdengdui
2025-08-13 16:10 ` Stephen Hemminger
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=079a2899-b2c4-4e50-b67a-eaff5b04e9ec@huawei.com \
--to=huangdengdui@huawei.com \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=fengchengwen@huawei.com \
--cc=lihuisong@huawei.com \
--cc=liuyonglong@huawei.com \
--cc=mb@smartsharesystems.com \
--cc=roretzla@linux.microsoft.com \
--cc=stephen@networkplumber.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.