From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: "Yang, Ziye" <ziye.yang@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: Ziye Yang <optimistyzy@gmail.com>
Subject: Re: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
Date: Mon, 10 Sep 2018 12:19:27 +0000 [thread overview]
Message-ID: <2601191342CEEE43887BDE71AB977258EA953AD7@irsmsx105.ger.corp.intel.com> (raw)
In-Reply-To: <FA6C2217B01E9D48A581BB486602101439AB0E30@shsmsx102.ccr.corp.intel.com>
> -----Original Message-----
> From: Yang, Ziye
> Sent: Monday, September 10, 2018 12:58 PM
> To: Ananyev, Konstantin <konstantin.ananyev@intel.com>; dev@dpdk.org
> Cc: Ziye Yang <optimistyzy@gmail.com>
> Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
>
> Hi Konstantin,
>
> The global logid is used to prevent the memory leak. If we do not do this, but directly free(logid), this pointer will still be used by
> openlog, which means that we cannot free the logid anymore. And by adding the global variable, it will not only solve the memory
> leak but also makes the openlog function still work.
But then I suppose you can avoid any dynamic memory allocation at all with something like that:
rte_eal_init(int argc, char **argv)
{
...
static char logid[PATH_MAX];
...
const char *p= strrchr(argv[0], '/');
snprintf(logid, sizeof(logid), "%s", p);
...
rte_eal_log_init(logid, ...);
?
Konstantin
>
>
> Best Regards
> Ziye Yang
>
> -----Original Message-----
> From: Ananyev, Konstantin
> Sent: Monday, September 10, 2018 7:55 PM
> To: Yang, Ziye <ziye.yang@intel.com>; dev@dpdk.org
> Cc: Ziye Yang <optimistyzy@gmail.com>
> Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
>
> Hi
>
> > -----Original Message-----
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Ziye Yang
> > Sent: Wednesday, September 5, 2018 6:39 AM
> > To: dev@dpdk.org
> > Cc: Ziye Yang <optimistyzy@gmail.com>
> > Subject: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak
> > issue of logid
> >
> > From: Ziye Yang <optimistyzy@gmail.com>
> >
> > This patch is used to fix the memory leak issue of logid.
> > We use the ASAN test in SPDK when intergrating DPDK and find this
> > memory leak issue.
> >
> > Signed-off-by: Ziye Yang <ziye.yang@intel.com>
> > ---
> > lib/librte_eal/linuxapp/eal/eal.c | 21 ++++++++++++++++++++-
> > lib/librte_eal/linuxapp/eal/eal_log.c | 11 ++++++++++-
> > 2 files changed, 30 insertions(+), 2 deletions(-)
> >
> > diff --git a/lib/librte_eal/linuxapp/eal/eal.c
> > b/lib/librte_eal/linuxapp/eal/eal.c
> > index e59ac65..e150200 100644
> > --- a/lib/librte_eal/linuxapp/eal/eal.c
> > +++ b/lib/librte_eal/linuxapp/eal/eal.c
> > @@ -793,7 +793,7 @@ static void rte_eal_init_alert(const char *msg)
> > int i, fctret, ret;
> > pthread_t thread_id;
> > static rte_atomic32_t run_once = RTE_ATOMIC32_INIT(0);
> > - const char *logid;
> > + char *logid;
> > char cpuset[RTE_CPU_AFFINITY_STR_LEN];
> > char thread_name[RTE_MAX_THREAD_NAME_LEN];
> >
> > @@ -812,6 +812,12 @@ static void rte_eal_init_alert(const char *msg)
> >
> > logid = strrchr(argv[0], '/');
> > logid = strdup(logid ? logid + 1: argv[0]);
> > + if (!logid) {
> > + rte_eal_init_alert("Cannot allocate memory for logid\n");
> > + rte_errno = ENOMEM;
> > + rte_atomic32_clear(&run_once);
> > + return -1;
> > + }
> >
> > thread_id = pthread_self();
> >
> > @@ -823,6 +829,8 @@ static void rte_eal_init_alert(const char *msg)
> > if (rte_eal_cpu_init() < 0) {
> > rte_eal_init_alert("Cannot detect lcores.");
> > rte_errno = ENOTSUP;
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -831,6 +839,7 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Invalid 'command line' arguments.");
> > rte_errno = EINVAL;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -838,12 +847,14 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot init plugins\n");
> > rte_errno = EINVAL;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > if (eal_option_device_parse()) {
> > rte_errno = ENODEV;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -851,6 +862,8 @@ static void rte_eal_init_alert(const char *msg)
> >
> > if (rte_eal_intr_init() < 0) {
> > rte_eal_init_alert("Cannot init interrupt-handling thread\n");
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -861,6 +874,8 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("failed to init mp channel\n");
> > if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
> > rte_errno = EFAULT;
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> > }
> > @@ -869,6 +884,7 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot scan the buses for devices\n");
> > rte_errno = ENODEV;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -893,6 +909,7 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot get hugepage information.");
> > rte_errno = EACCES;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> > }
> > @@ -919,8 +936,10 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot init logging.");
> > rte_errno = ENOMEM;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> > + free(logid);
> >
> > #ifdef VFIO_PRESENT
> > if (rte_eal_vfio_setup() < 0) {
> > diff --git a/lib/librte_eal/linuxapp/eal/eal_log.c
> > b/lib/librte_eal/linuxapp/eal/eal_log.c
> > index 9d02ddd..076b1b9 100644
> > --- a/lib/librte_eal/linuxapp/eal/eal_log.c
> > +++ b/lib/librte_eal/linuxapp/eal/eal_log.c
> > @@ -19,6 +19,9 @@
> >
> > #include "eal_private.h"
> >
> > +#define DEFAULT_LOG_ID_LEN 128
> > +char g_logid[DEFAULT_LOG_ID_LEN];
> > +
> > /*
> > * default log function
> > */
> > @@ -49,12 +52,18 @@
> > rte_eal_log_init(const char *id, int facility) {
> > FILE *log_stream;
> > + int str_len;
> >
> > log_stream = fopencookie(NULL, "w+", console_log_func);
> > if (log_stream == NULL)
> > return -1;
> >
> > - openlog(id, LOG_NDELAY | LOG_PID, facility);
> > + str_len = strlen(id);
> > + if (str_len >= DEFAULT_LOG_ID_LEN)
> > + return -1;
> > +
> > + strncpy(g_logid, id, str_len);
> > + openlog(g_logid, LOG_NDELAY | LOG_PID, facility);
>
>
> Could you explain what is the point here to create a global here and pass pointer to it to openlog()?
> Konstantin
>
> >
> > eal_log_set_default(log_stream);
> >
> > --
> > 1.9.3
next prev parent reply other threads:[~2018-09-10 12:19 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-05 5:39 Ziye Yang
2018-09-10 11:54 ` Ananyev, Konstantin
2018-09-10 11:58 ` Yang, Ziye
2018-09-10 12:19 ` Ananyev, Konstantin [this message]
2018-09-10 12:23 ` Yang, Ziye
2018-09-10 13:46 ` [dpdk-dev] [PATCH v6] " Ziye Yang
2018-09-11 1:27 ` [dpdk-dev] [PATCH v7] " Ziye Yang
2018-09-11 10:09 ` Ananyev, Konstantin
2018-09-11 13:47 ` Aaron Conole
2018-09-11 14:06 ` Ananyev, Konstantin
2018-09-11 15:27 ` Aaron Conole
2018-09-13 13:28 ` Aaron Conole
2018-09-12 1:31 ` [dpdk-dev] [PATCH v8] " Ziye Yang
2018-10-16 15:33 ` Ferruh Yigit
2018-10-22 8:00 ` Thomas Monjalon
2018-10-28 10:41 ` 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=2601191342CEEE43887BDE71AB977258EA953AD7@irsmsx105.ger.corp.intel.com \
--to=konstantin.ananyev@intel.com \
--cc=dev@dpdk.org \
--cc=optimistyzy@gmail.com \
--cc=ziye.yang@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).