DPDK patches and discussions
 help / color / mirror / Atom feed
* [dpdk-dev] [PATCH v1] test: fix ipsec unit test segfault
@ 2020-03-20  8:05 Ruifeng Wang
  2020-03-23 12:21 ` Ananyev, Konstantin
  0 siblings, 1 reply; 3+ messages in thread
From: Ruifeng Wang @ 2020-03-20  8:05 UTC (permalink / raw)
  To: bernard.iremonger, konstantin.ananyev
  Cc: dev, gavin.hu, honnappa.nagarahalli, juraj.linkes, nd,
	Ruifeng Wang, stable

Segfault was observed when running ipsec unit test:

 + TestCase [10] : test_ipsec_replay_inb_repeat_null_null_wrapper
                   succeeded
 + TestCase [11] : test_ipsec_replay_inb_inside_burst_null_null_wrapper
                   succeeded
 + TestCase [12] : test_ipsec_crypto_inb_burst_2sa_null_null_wrapper
                   succeeded
 + TestCase [13] : test_ipsec_crypto_inb_burst_2sa_4grp_null_null_wrapper
                   succeeded
Segmentation fault

Data corruption happens due to incorrect destroy of session. Security
session needs process different from crypto session.

Destroy corresponding sessions according to different security actions.

Fixes: 05fe65eb66b2 ("test/ipsec: introduce functional test")
Cc: stable@dpdk.org

Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
---
 app/test/test_ipsec.c | 33 ++++++++++++++++++++++++++++++---
 1 file changed, 30 insertions(+), 3 deletions(-)

diff --git a/app/test/test_ipsec.c b/app/test/test_ipsec.c
index 9e1447293..79d00d7e0 100644
--- a/app/test/test_ipsec.c
+++ b/app/test/test_ipsec.c
@@ -1167,6 +1167,34 @@ test_ipsec_dump_buffers(struct ipsec_unitest_params *ut_params, int i)
 	}
 }
 
+static void
+destroy_dummy_sec_session(struct ipsec_unitest_params *ut,
+	uint32_t j)
+{
+	rte_security_session_destroy(&dummy_sec_ctx,
+					ut->ss[j].security.ses);
+	ut->ss[j].security.ctx = NULL;
+}
+
+static void
+destroy_crypto_session(struct ipsec_unitest_params *ut,
+	uint8_t crypto_dev, uint32_t j)
+{
+	rte_cryptodev_sym_session_clear(crypto_dev, ut->ss[j].crypto.ses);
+	rte_cryptodev_sym_session_free(ut->ss[j].crypto.ses);
+	memset(&ut->ss[j], 0, sizeof(ut->ss[j]));
+}
+
+static void
+destroy_session(struct ipsec_unitest_params *ut,
+	uint8_t crypto_dev, uint32_t j)
+{
+	if (ut->ss[j].type == RTE_SECURITY_ACTION_TYPE_NONE)
+		return destroy_crypto_session(ut, crypto_dev, j);
+	else
+		return destroy_dummy_sec_session(ut, j);
+}
+
 static void
 destroy_sa(uint32_t j)
 {
@@ -1175,9 +1203,8 @@ destroy_sa(uint32_t j)
 
 	rte_ipsec_sa_fini(ut->ss[j].sa);
 	rte_free(ut->ss[j].sa);
-	rte_cryptodev_sym_session_clear(ts->valid_dev, ut->ss[j].crypto.ses);
-	rte_cryptodev_sym_session_free(ut->ss[j].crypto.ses);
-	memset(&ut->ss[j], 0, sizeof(ut->ss[j]));
+
+	destroy_session(ut, ts->valid_dev, j);
 }
 
 static int
-- 
2.17.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [dpdk-dev] [PATCH v1] test: fix ipsec unit test segfault
  2020-03-20  8:05 [dpdk-dev] [PATCH v1] test: fix ipsec unit test segfault Ruifeng Wang
@ 2020-03-23 12:21 ` Ananyev, Konstantin
  2020-03-25 19:03   ` Akhil Goyal
  0 siblings, 1 reply; 3+ messages in thread
From: Ananyev, Konstantin @ 2020-03-23 12:21 UTC (permalink / raw)
  To: Ruifeng Wang, Iremonger, Bernard
  Cc: dev, gavin.hu, honnappa.nagarahalli, juraj.linkes, nd, stable


> 
> Segfault was observed when running ipsec unit test:
> 
>  + TestCase [10] : test_ipsec_replay_inb_repeat_null_null_wrapper
>                    succeeded
>  + TestCase [11] : test_ipsec_replay_inb_inside_burst_null_null_wrapper
>                    succeeded
>  + TestCase [12] : test_ipsec_crypto_inb_burst_2sa_null_null_wrapper
>                    succeeded
>  + TestCase [13] : test_ipsec_crypto_inb_burst_2sa_4grp_null_null_wrapper
>                    succeeded
> Segmentation fault
> 
> Data corruption happens due to incorrect destroy of session. Security
> session needs process different from crypto session.
> 
> Destroy corresponding sessions according to different security actions.
> 
> Fixes: 05fe65eb66b2 ("test/ipsec: introduce functional test")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
> Reviewed-by: Phil Yang <phil.yang@arm.com>
> Reviewed-by: Gavin Hu <gavin.hu@arm.com>
> ---
>  app/test/test_ipsec.c | 33 ++++++++++++++++++++++++++++++---
>  1 file changed, 30 insertions(+), 3 deletions(-)
> 
> diff --git a/app/test/test_ipsec.c b/app/test/test_ipsec.c
> index 9e1447293..79d00d7e0 100644
> --- a/app/test/test_ipsec.c
> +++ b/app/test/test_ipsec.c
> @@ -1167,6 +1167,34 @@ test_ipsec_dump_buffers(struct ipsec_unitest_params *ut_params, int i)
>  	}
>  }
> 
> +static void
> +destroy_dummy_sec_session(struct ipsec_unitest_params *ut,
> +	uint32_t j)
> +{
> +	rte_security_session_destroy(&dummy_sec_ctx,
> +					ut->ss[j].security.ses);
> +	ut->ss[j].security.ctx = NULL;
> +}
> +
> +static void
> +destroy_crypto_session(struct ipsec_unitest_params *ut,
> +	uint8_t crypto_dev, uint32_t j)
> +{
> +	rte_cryptodev_sym_session_clear(crypto_dev, ut->ss[j].crypto.ses);
> +	rte_cryptodev_sym_session_free(ut->ss[j].crypto.ses);
> +	memset(&ut->ss[j], 0, sizeof(ut->ss[j]));
> +}
> +
> +static void
> +destroy_session(struct ipsec_unitest_params *ut,
> +	uint8_t crypto_dev, uint32_t j)
> +{
> +	if (ut->ss[j].type == RTE_SECURITY_ACTION_TYPE_NONE)
> +		return destroy_crypto_session(ut, crypto_dev, j);
> +	else
> +		return destroy_dummy_sec_session(ut, j);
> +}
> +
>  static void
>  destroy_sa(uint32_t j)
>  {
> @@ -1175,9 +1203,8 @@ destroy_sa(uint32_t j)
> 
>  	rte_ipsec_sa_fini(ut->ss[j].sa);
>  	rte_free(ut->ss[j].sa);
> -	rte_cryptodev_sym_session_clear(ts->valid_dev, ut->ss[j].crypto.ses);
> -	rte_cryptodev_sym_session_free(ut->ss[j].crypto.ses);
> -	memset(&ut->ss[j], 0, sizeof(ut->ss[j]));
> +
> +	destroy_session(ut, ts->valid_dev, j);
>  }
> 
>  static int
> --

Good catch, thanks for the fix.
As a nit caption should probably be: "test/ipsec: fix ..."

Tested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>

> 2.17.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [dpdk-dev] [PATCH v1] test: fix ipsec unit test segfault
  2020-03-23 12:21 ` Ananyev, Konstantin
@ 2020-03-25 19:03   ` Akhil Goyal
  0 siblings, 0 replies; 3+ messages in thread
From: Akhil Goyal @ 2020-03-25 19:03 UTC (permalink / raw)
  To: Ananyev, Konstantin, Ruifeng Wang, Iremonger, Bernard
  Cc: dev, gavin.hu, honnappa.nagarahalli, juraj.linkes, nd, stable

> >
> > Segfault was observed when running ipsec unit test:
> >
> >  + TestCase [10] : test_ipsec_replay_inb_repeat_null_null_wrapper
> >                    succeeded
> >  + TestCase [11] : test_ipsec_replay_inb_inside_burst_null_null_wrapper
> >                    succeeded
> >  + TestCase [12] : test_ipsec_crypto_inb_burst_2sa_null_null_wrapper
> >                    succeeded
> >  + TestCase [13] : test_ipsec_crypto_inb_burst_2sa_4grp_null_null_wrapper
> >                    succeeded
> > Segmentation fault
> >
> > Data corruption happens due to incorrect destroy of session. Security
> > session needs process different from crypto session.
> >
> > Destroy corresponding sessions according to different security actions.
> >
> > Fixes: 05fe65eb66b2 ("test/ipsec: introduce functional test")
> > Cc: stable@dpdk.org
> >
> > Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
> > Reviewed-by: Phil Yang <phil.yang@arm.com>
> > Reviewed-by: Gavin Hu <gavin.hu@arm.com>
> > ---
> 
> Good catch, thanks for the fix.
> As a nit caption should probably be: "test/ipsec: fix ..."

Fixed
Applied to dpdk-next-crypto

Thanks.
> 
> Tested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> 
> > 2.17.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2020-03-25 19:04 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-03-20  8:05 [dpdk-dev] [PATCH v1] test: fix ipsec unit test segfault Ruifeng Wang
2020-03-23 12:21 ` Ananyev, Konstantin
2020-03-25 19:03   ` Akhil Goyal

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).