DPDK patches and discussions
 help / color / mirror / Atom feed
From: Akhil Goyal <akhil.goyal@nxp.com>
To: Ciara Power <ciara.power@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: "roy.fan.zhang@intel.com" <roy.fan.zhang@intel.com>,
	Declan Doherty <declan.doherty@intel.com>
Subject: Re: [dpdk-dev] [PATCH] test/crypto: fix null dereference for crypto op
Date: Mon, 2 Nov 2020 19:55:21 +0000	[thread overview]
Message-ID: <VI1PR04MB31688684FA1CB1D1753E4A1EE6100@VI1PR04MB3168.eurprd04.prod.outlook.com> (raw)
In-Reply-To: <20201030131126.222553-1-ciara.power@intel.com>

> Subject: [dpdk-dev] [PATCH] test/crypto: fix null dereference for crypto op
> 
> In two test cases, the op value is set by the return of the
> process_crypto_request function, which may be NULL. The op->status
> value was checked afterwards, which was causing a dereference issue.
> 
> To fix this, a temporary op variable is used to hold the return
> from the process_crypto_request function, so the original op->status
> can be checked after the possible NULL return value.
> The original op value is then set to hold the temporary op value.
> 
> Coverity issue: 363465
> Coverity issue: 363452
> Fixes: 4868f6591c6f ("test/crypto: add cases for raw datapath API")
> Cc: roy.fan.zhang@intel.com
> 
> Signed-off-by: Ciara Power <ciara.power@intel.com>
> ---
>  app/test/test_cryptodev.c | 14 ++++++++------
>  1 file changed, 8 insertions(+), 6 deletions(-)
> 
> diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
> index 0fed124d3a..ce8bcd1d4f 100644
> --- a/app/test/test_cryptodev.c
> +++ b/app/test/test_cryptodev.c
> @@ -6676,6 +6676,7 @@ test_mixed_auth_cipher(const struct
> mixed_cipher_auth_test_data *tdata,
>  	unsigned int ciphertext_len;
> 
>  	struct rte_cryptodev_info dev_info;
> +	struct rte_crypto_op *op;
> 
>  	/* Check if device supports particular algorithms separately */
>  	if (test_mixed_check_if_unsupported(tdata))
> @@ -6771,17 +6772,17 @@ test_mixed_auth_cipher(const struct
> mixed_cipher_auth_test_data *tdata,
>  	if (retval < 0)
>  		return retval;
> 
> -	ut_params->op = process_crypto_request(ts_params->valid_devs[0],
> -			ut_params->op);
> +	op = process_crypto_request(ts_params->valid_devs[0], ut_params-
> >op);
> 
>  	/* Check if the op failed because the device doesn't */
>  	/* support this particular combination of algorithms */
> -	if (ut_params->op == NULL && ut_params->op->status ==
> +	if (op == NULL && ut_params->op->status ==
>  			RTE_CRYPTO_OP_STATUS_INVALID_SESSION) {
>  		printf("Device doesn't support this mixed combination. "
>  				"Test Skipped.\n");
>  		return -ENOTSUP;
>  	}

I believe the original check was also good. But instead of 
if (ut_params->op == NULL && ut_params->op->status == RTE_CRYPTO_OP_STATUS_INVALID_SESSION)
it should be
if (ut_params->op == NULL || ut_params->op->status == RTE_CRYPTO_OP_STATUS_INVALID_SESSION)

In this way the coverity should not raise an issue for this. What say?

> +	ut_params->op = op;
> 
>  	TEST_ASSERT_NOT_NULL(ut_params->op, "failed to retrieve obuf");
> 
> @@ -6872,6 +6873,7 @@ test_mixed_auth_cipher_sgl(const struct
> mixed_cipher_auth_test_data *tdata,
>  	uint8_t digest_buffer[10000];
> 
>  	struct rte_cryptodev_info dev_info;
> +	struct rte_crypto_op *op;
> 
>  	/* Check if device supports particular algorithms */
>  	if (test_mixed_check_if_unsupported(tdata))
> @@ -6976,17 +6978,17 @@ test_mixed_auth_cipher_sgl(const struct
> mixed_cipher_auth_test_data *tdata,
>  	if (retval < 0)
>  		return retval;
> 
> -	ut_params->op = process_crypto_request(ts_params->valid_devs[0],
> -			ut_params->op);
> +	op = process_crypto_request(ts_params->valid_devs[0], ut_params-
> >op);
> 
>  	/* Check if the op failed because the device doesn't */
>  	/* support this particular combination of algorithms */
> -	if (ut_params->op == NULL && ut_params->op->status ==
> +	if (op == NULL && ut_params->op->status ==
>  			RTE_CRYPTO_OP_STATUS_INVALID_SESSION) {
>  		printf("Device doesn't support this mixed combination. "
>  				"Test Skipped.\n");
>  		return -ENOTSUP;
>  	}
> +	ut_params->op = op;
> 
>  	TEST_ASSERT_NOT_NULL(ut_params->op, "failed to retrieve obuf");
> 
> --
> 2.25.1


  reply	other threads:[~2020-11-02 19:55 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-30 13:11 Ciara Power
2020-11-02 19:55 ` Akhil Goyal [this message]
2020-11-04  7:16   ` Akhil Goyal

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=VI1PR04MB31688684FA1CB1D1753E4A1EE6100@VI1PR04MB3168.eurprd04.prod.outlook.com \
    --to=akhil.goyal@nxp.com \
    --cc=ciara.power@intel.com \
    --cc=declan.doherty@intel.com \
    --cc=dev@dpdk.org \
    --cc=roy.fan.zhang@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).