From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
To: "Yu, DapengX" <dapengx.yu@intel.com>
Cc: Cristian Dumitrescu <cristian.dumitrescu@intel.com>,
Jasvinder Singh <jasvinder.singh@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v2] net/softnic: fix memory leak in parsing arguments
Date: Tue, 13 Jul 2021 13:29:15 +0300 [thread overview]
Message-ID: <e5597159-6c75-3dab-c8d8-b251ff0ee485@oktetlabs.ru> (raw)
In-Reply-To: <PH0PR11MB512563E3EA51977861DF77CF8C149@PH0PR11MB5125.namprd11.prod.outlook.com>
Adding list and maintainers back.
On 7/13/21 1:18 PM, Yu, DapengX wrote:
>> -----Original Message-----
>> From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
>> Sent: Tuesday, July 13, 2021 6:11 PM
>> To: Yu, DapengX <dapengx.yu@intel.com>; Singh, Jasvinder
>> <jasvinder.singh@intel.com>; Dumitrescu, Cristian
>> <cristian.dumitrescu@intel.com>
>> Cc: dev@dpdk.org; stable@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH v2] net/softnic: fix memory leak in parsing
>> arguments
>>
>> On 7/13/21 1:08 PM, dapengx.yu@intel.com wrote:
>>> From: Dapeng Yu <dapengx.yu@intel.com>
>>>
>>> In function pmd_parse_args(), firmware path is duplicated from device
>>> arguments as character string, but is never freed, which cause memory
>>> leak.
>>>
>>> This patch changes the type of firmware member of struct pmd_params to
>>> character array, to make memory resource release unnecessary, and
>>> changes the type of name member to character array, to keep the
>>> consistency of character string handling in struct pmd_params.
>>>
>>> Fixes: 7e68bc20f8c8 ("net/softnic: restructure")
>>> Cc: stable@dpdk.org
>>>
>>> Signed-off-by: Dapeng Yu <dapengx.yu@intel.com>
>>> ---
>>> V2:
>>> * improve the patch according to maintainer's comment:
>>> rte_strscpy() is the best option here.
>>> ---
>>> drivers/net/softnic/rte_eth_softnic.c | 9 ++++++---
>>> drivers/net/softnic/rte_eth_softnic_internals.h | 4 ++--
>>> 2 files changed, 8 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/net/softnic/rte_eth_softnic.c
>>> b/drivers/net/softnic/rte_eth_softnic.c
>>> index f64023256d..8a49e83dce 100644
>>> --- a/drivers/net/softnic/rte_eth_softnic.c
>>> +++ b/drivers/net/softnic/rte_eth_softnic.c
>>> @@ -440,6 +440,7 @@ pmd_parse_args(struct pmd_params *p, const char
>>> *params) {
>>> struct rte_kvargs *kvlist;
>>> int ret = 0;
>>> + char *firmware = NULL;
>>>
>>> kvlist = rte_kvargs_parse(params, pmd_valid_args);
>>> if (kvlist == NULL)
>>> @@ -447,7 +448,7 @@ pmd_parse_args(struct pmd_params *p, const char
>>> *params)
>>>
>>> /* Set default values */
>>> memset(p, 0, sizeof(*p));
>>> - p->firmware = SOFTNIC_FIRMWARE;
>>> + rte_strscpy(p->firmware, SOFTNIC_FIRMWARE, sizeof(p-
>>> firmware));
>>
>> I still don't understand why return value is not checked here and in similar
>> cases below.
> If the return value is not checked here, and the dst path string is not complete when the path string is too long, softnic_cli_script_process() will prompt error.
> So the fix will not cause more negative impact.
> And for the similar case below, the src and dst string length is same always, so no negative impact too.
I see. For me it sounds a bit fragile, but not critical.
Anyway since I'm not 100% sure I'll wait for maintainers
review.
>>
>>> p->cpu_id = SOFTNIC_CPU_ID;
>>> p->sc = SOFTNIC_SC;
>>> p->tm.n_queues = SOFTNIC_TM_N_QUEUES; @@ -468,10 +469,12
>> @@
>>> pmd_parse_args(struct pmd_params *p, const char *params)
>>> /* Firmware script (optional) */
>>> if (rte_kvargs_count(kvlist, PMD_PARAM_FIRMWARE) == 1) {
>>> ret = rte_kvargs_process(kvlist, PMD_PARAM_FIRMWARE,
>>> - &get_string, &p->firmware);
>>> + &get_string, &firmware);
>>> if (ret < 0)
>>> goto out_free;
>>> }
>>> + rte_strscpy(p->firmware, firmware, sizeof(p->firmware));
>>> + free(firmware);
>>>
>>> /* Connection listening port (optional) */
>>> if (rte_kvargs_count(kvlist, PMD_PARAM_CONN_PORT) == 1) { @@
>> -621,7
>>> +624,7 @@ pmd_probe(struct rte_vdev_device *vdev)
>>> if (status)
>>> return status;
>>>
>>> - p.name = name;
>>> + rte_strscpy(p.name, name, sizeof(p.name));
>>>
>>> /* Allocate and initialize soft ethdev private data */
>>> dev_private = pmd_init(&p);
>>> diff --git a/drivers/net/softnic/rte_eth_softnic_internals.h
>>> b/drivers/net/softnic/rte_eth_softnic_internals.h
>>> index 1b3186ef0b..a13d67b7c1 100644
>>> --- a/drivers/net/softnic/rte_eth_softnic_internals.h
>>> +++ b/drivers/net/softnic/rte_eth_softnic_internals.h
>>> @@ -34,8 +34,8 @@
>>> */
>>>
>>> struct pmd_params {
>>> - const char *name;
>>> - const char *firmware;
>>> + char name[RTE_DEV_NAME_MAX_LEN];
>>> + char firmware[PATH_MAX];
>>> uint16_t conn_port;
>>> uint32_t cpu_id;
>>> int sc; /**< Service cores. */
>>>
>
next prev parent reply other threads:[~2021-07-13 10:29 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-08 8:44 [dpdk-dev] [PATCH] " dapengx.yu
2021-07-13 8:44 ` Andrew Rybchenko
2021-07-13 10:08 ` [dpdk-dev] [PATCH v2] " dapengx.yu
2021-07-13 10:10 ` Andrew Rybchenko
[not found] ` <PH0PR11MB512563E3EA51977861DF77CF8C149@PH0PR11MB5125.namprd11.prod.outlook.com>
2021-07-13 10:29 ` Andrew Rybchenko [this message]
2021-07-14 5:47 ` [dpdk-dev] [PATCH v3] " dapengx.yu
2021-07-14 11:07 ` Singh, Jasvinder
2021-07-15 1:42 ` Yu, DapengX
2021-07-15 5:38 ` [dpdk-dev] [PATCH v4] " dapengx.yu
2021-07-16 12:04 ` Singh, Jasvinder
2021-07-23 8:11 ` Thomas Monjalon
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=e5597159-6c75-3dab-c8d8-b251ff0ee485@oktetlabs.ru \
--to=andrew.rybchenko@oktetlabs.ru \
--cc=cristian.dumitrescu@intel.com \
--cc=dapengx.yu@intel.com \
--cc=dev@dpdk.org \
--cc=jasvinder.singh@intel.com \
/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).