DPDK patches and discussions
 help / color / mirror / Atom feed
From: Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>
To: Lukasz Wojciechowski <l.wojciechow@partner.samsung.com>,
	"dev@dpdk.org" <dev@dpdk.org>,
	"john.mcnamara@intel.com" <john.mcnamara@intel.com>
Cc: Ruifeng Wang <Ruifeng.Wang@arm.com>,
	"juraj.linkes@pantheon.tech" <juraj.linkes@pantheon.tech>,
	"david.marchand@redhat.com" <david.marchand@redhat.com>,
	nd <nd@arm.com>, "stable@dpdk.org" <stable@dpdk.org>,
	Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>,
	nd <nd@arm.com>
Subject: Re: [dpdk-dev] [PATCH 1/2] test/rcu: fix array subscript is above array bounds
Date: Tue, 20 Oct 2020 16:26:44 +0000	[thread overview]
Message-ID: <AM8PR08MB5810C4B08673FA071091719D981F0@AM8PR08MB5810.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <791ffc96-8d87-5783-c596-92e21cdda913@partner.samsung.com>

<snip>
Hi Lukasz,

> 
> Hi Honnappa,
> 
> I verified building and testing and all the warnings/errors disappear for
> RTE_MAX_LCORE >= 2 and tests pass.
Thank you for testing this.

> 
> I wonder, if it is possible to set RTE_MAX_LCORE = 1 ?
I thought, we would need 2 cores minimum, one for main and the other for worker.
I compiled now with 1 core. I see more errors than what you are seeing. I am seeing errors in test cases for bbdev, hash, lpm as well. Not sure if it is worth fixing them.

> In such case there are still few places with array bounds exceedings:
> Compiling C object 'app/test/3062f5d@@dpdk-test@exe/test_rcu_qsbr.c.o'.
> ../app/test/test_rcu_qsbr.c: In function ‘test_rcu_qsbr_check_reader’:
> ../app/test/test_rcu_qsbr.c:319:24: warning: array subscript is above array
> bounds [-Warray-bounds]
>          enabled_core_ids[i]);
>          ~~~~~~~~~~~~~~~~^~~
> ../app/test/test_rcu_qsbr.c: In function ‘test_rcu_qsbr_main’:
> ../app/test/test_rcu_qsbr.c:946:2: warning: array subscript is above array
> bounds [-Warray-bounds]
>    rte_rcu_qsbr_init(t[1], RTE_MAX_LCORE);
>    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ../app/test/test_rcu_qsbr.c:954:3: warning: array subscript is above array
> bounds [-Warray-bounds]
>     rte_rcu_qsbr_thread_register(t[1], enabled_core_ids[i]);
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ../app/test/test_rcu_qsbr.c:957:2: warning: array subscript is above array
> bounds [-Warray-bounds]
>    rte_rcu_qsbr_dump(stdout, t[1]);
>    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ../app/test/test_rcu_qsbr.c:486:53: warning: array subscript is above array
> bounds [-Warray-bounds]
>    rte_rcu_qsbr_thread_register(t[0], enabled_core_ids[1]);
>                                       ~~~~~~~~~~~~~~~~^~~ ...and few more in other files.
> 
> 
> Best regards
> 
> Lukasz
> 
> W dniu 16.10.2020 o 08:03, Honnappa Nagarahalli pisze:
> > When RTE_MAX_LCORE value is small, following compiler errors are
> > observed.
> >
> > ../app/test/test_rcu_qsbr.c:296:54: error: iteration 2 invokes
> > undefined behavior [-Werror=aggressive-loop-optimizations]
> >
> > ../app/test/test_rcu_qsbr.c:315:55: error: array subscript is above
> > array bounds [-Werror=array-bounds]
> >
> > Fixes: b87089b0bb19 ("test/rcu: add API and functional tests")
> > Cc: stable@dpdk.org
> >
> > Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
> > Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
> > ---
> >   app/test/test_rcu_qsbr.c | 56 +++++++++++++++++++++++----------------
> -
> >   1 file changed, 33 insertions(+), 23 deletions(-)
> >
> > diff --git a/app/test/test_rcu_qsbr.c b/app/test/test_rcu_qsbr.c index
> > 0a9e5ecd1..848a68092 100644
> > --- a/app/test/test_rcu_qsbr.c
> > +++ b/app/test/test_rcu_qsbr.c
> > @@ -286,13 +286,13 @@ static int
> >   test_rcu_qsbr_start(void)
> >   {
> >   	uint64_t token;
> > -	int i;
> > +	unsigned int i;
> >
> >   	printf("\nTest rte_rcu_qsbr_start()\n");
> >
> >   	rte_rcu_qsbr_init(t[0], RTE_MAX_LCORE);
> >
> > -	for (i = 0; i < 3; i++)
> > +	for (i = 0; i < num_cores; i++)
> >   		rte_rcu_qsbr_thread_register(t[0], enabled_core_ids[i]);
> >
> >   	token = rte_rcu_qsbr_start(t[0]);
> > @@ -306,14 +306,18 @@ test_rcu_qsbr_check_reader(void *arg)
> >   {
> >   	struct rte_rcu_qsbr *temp;
> >   	uint8_t read_type = (uint8_t)((uintptr_t)arg);
> > +	unsigned int i;
> >
> >   	temp = t[read_type];
> >
> >   	/* Update quiescent state counter */
> > -	rte_rcu_qsbr_quiescent(temp, enabled_core_ids[0]);
> > -	rte_rcu_qsbr_quiescent(temp, enabled_core_ids[1]);
> > -	rte_rcu_qsbr_thread_unregister(temp, enabled_core_ids[2]);
> > -	rte_rcu_qsbr_quiescent(temp, enabled_core_ids[3]);
> > +	for (i = 0; i < num_cores; i++) {
> > +		if (i % 2 == 0)
> > +			rte_rcu_qsbr_quiescent(temp, enabled_core_ids[i]);
> > +		else
> > +			rte_rcu_qsbr_thread_unregister(temp,
> > +							enabled_core_ids[i]);
> > +	}
> >   	return 0;
> >   }
> >
> > @@ -324,7 +328,8 @@ test_rcu_qsbr_check_reader(void *arg)
> >   static int
> >   test_rcu_qsbr_check(void)
> >   {
> > -	int i, ret;
> > +	int ret;
> > +	unsigned int i;
> >   	uint64_t token;
> >
> >   	printf("\nTest rte_rcu_qsbr_check()\n"); @@ -342,7 +347,7 @@
> > test_rcu_qsbr_check(void)
> >   	ret = rte_rcu_qsbr_check(t[0], token, true);
> >   	TEST_RCU_QSBR_RETURN_IF_ERROR((ret == 0), "Blocking QSBR
> check");
> >
> > -	for (i = 0; i < 3; i++)
> > +	for (i = 0; i < num_cores; i++)
> >   		rte_rcu_qsbr_thread_register(t[0], enabled_core_ids[i]);
> >
> >   	ret = rte_rcu_qsbr_check(t[0], token, false); @@ -357,7 +362,7 @@
> > test_rcu_qsbr_check(void)
> >   	/* Threads are offline, hence this should pass */
> >   	TEST_RCU_QSBR_RETURN_IF_ERROR((ret == 0), "Non-blocking QSBR
> > check");
> >
> > -	for (i = 0; i < 3; i++)
> > +	for (i = 0; i < num_cores; i++)
> >   		rte_rcu_qsbr_thread_unregister(t[0], enabled_core_ids[i]);
> >
> >   	ret = rte_rcu_qsbr_check(t[0], token, true); @@ -365,7 +370,7 @@
> > test_rcu_qsbr_check(void)
> >
> >   	rte_rcu_qsbr_init(t[0], RTE_MAX_LCORE);
> >
> > -	for (i = 0; i < 4; i++)
> > +	for (i = 0; i < num_cores; i++)
> >   		rte_rcu_qsbr_thread_register(t[0], enabled_core_ids[i]);
> >
> >   	token = rte_rcu_qsbr_start(t[0]);
> > @@ -928,7 +933,7 @@ test_rcu_qsbr_dq_functional(int32_t size, int32_t
> esize, uint32_t flags)
> >   static int
> >   test_rcu_qsbr_dump(void)
> >   {
> > -	int i;
> > +	unsigned int i;
> >
> >   	printf("\nTest rte_rcu_qsbr_dump()\n");
> >
> > @@ -945,7 +950,7 @@ test_rcu_qsbr_dump(void)
> >
> >   	rte_rcu_qsbr_thread_register(t[0], enabled_core_ids[0]);
> >
> > -	for (i = 1; i < 3; i++)
> > +	for (i = 1; i < num_cores; i++)
> >   		rte_rcu_qsbr_thread_register(t[1], enabled_core_ids[i]);
> >
> >   	rte_rcu_qsbr_dump(stdout, t[0]);
> > @@ -1095,7 +1100,7 @@ test_rcu_qsbr_sw_sv_3qs(void)
> >   {
> >   	uint64_t token[3];
> >   	uint32_t c;
> > -	int i;
> > +	int i, num_readers;
> >   	int32_t pos[3];
> >
> >   	writer_done = 0;
> > @@ -1118,7 +1123,11 @@ test_rcu_qsbr_sw_sv_3qs(void)
> >   	thread_info[0].ih = 0;
> >
> >   	/* Reader threads are launched */
> > -	for (i = 0; i < 4; i++)
> > +	/* Keep the number of reader threads low to reduce
> > +	 * the execution time.
> > +	 */
> > +	num_readers = num_cores < 4 ? num_cores : 4;
> > +	for (i = 0; i < num_readers; i++)
> >   		rte_eal_remote_launch(test_rcu_qsbr_reader,
> &thread_info[0],
> >   					enabled_core_ids[i]);
> >
> > @@ -1151,7 +1160,7 @@ test_rcu_qsbr_sw_sv_3qs(void)
> >
> >   	/* Check the quiescent state status */
> >   	rte_rcu_qsbr_check(t[0], token[0], true);
> > -	for (i = 0; i < 4; i++) {
> > +	for (i = 0; i < num_readers; i++) {
> >   		c = hash_data[0][0][enabled_core_ids[i]];
> >   		if (c != COUNTER_VALUE && c != 0) {
> >   			printf("Reader lcore %d did not complete #0 = %d\n",
> @@ -1169,7
> > +1178,7 @@ test_rcu_qsbr_sw_sv_3qs(void)
> >
> >   	/* Check the quiescent state status */
> >   	rte_rcu_qsbr_check(t[0], token[1], true);
> > -	for (i = 0; i < 4; i++) {
> > +	for (i = 0; i < num_readers; i++) {
> >   		c = hash_data[0][3][enabled_core_ids[i]];
> >   		if (c != COUNTER_VALUE && c != 0) {
> >   			printf("Reader lcore %d did not complete #3 = %d\n",
> @@ -1187,7
> > +1196,7 @@ test_rcu_qsbr_sw_sv_3qs(void)
> >
> >   	/* Check the quiescent state status */
> >   	rte_rcu_qsbr_check(t[0], token[2], true);
> > -	for (i = 0; i < 4; i++) {
> > +	for (i = 0; i < num_readers; i++) {
> >   		c = hash_data[0][6][enabled_core_ids[i]];
> >   		if (c != COUNTER_VALUE && c != 0) {
> >   			printf("Reader lcore %d did not complete #6 = %d\n",
> @@ -1206,7
> > +1215,7 @@ test_rcu_qsbr_sw_sv_3qs(void)
> >   	writer_done = 1;
> >
> >   	/* Wait and check return value from reader threads */
> > -	for (i = 0; i < 4; i++)
> > +	for (i = 0; i < num_readers; i++)
> >   		if (rte_eal_wait_lcore(enabled_core_ids[i]) < 0)
> >   			goto error;
> >   	rte_hash_free(h[0]);
> > @@ -1236,6 +1245,12 @@ test_rcu_qsbr_mw_mv_mqs(void)
> >   	unsigned int i, j;
> >   	unsigned int test_cores;
> >
> > +	if (RTE_MAX_LCORE < 5 || num_cores < 4) {
> > +		printf("Not enough cores for %s, expecting at least 5\n",
> > +			__func__);
> > +		return TEST_SKIPPED;
> > +	}
> > +
> >   	writer_done = 0;
> >   	test_cores = num_cores / 4;
> >   	test_cores = test_cores * 4;
> > @@ -1321,11 +1336,6 @@ test_rcu_qsbr_main(void)
> >   {
> >   	uint16_t core_id;
> >
> > -	if (rte_lcore_count() < 5) {
> > -		printf("Not enough cores for rcu_qsbr_autotest, expecting at
> least 5\n");
> > -		return TEST_SKIPPED;
> > -	}
> > -
> >   	num_cores = 0;
> >   	RTE_LCORE_FOREACH_SLAVE(core_id) {
> >   		enabled_core_ids[num_cores] = core_id;
> 
> --
> Lukasz Wojciechowski
> Principal Software Engineer
> 
> Samsung R&D Institute Poland
> Samsung Electronics
> Office +48 22 377 88 25
> l.wojciechow@partner.samsung.com


  reply	other threads:[~2020-10-20 16:27 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20201016060420eucas1p12f301a94eb4b4d19a9ced5c5cbd59c77@eucas1p1.samsung.com>
2020-10-16  6:03 ` Honnappa Nagarahalli
2020-10-16  6:03   ` [dpdk-dev] [PATCH 2/2] examples/performance-thread: fix undef behavior Honnappa Nagarahalli
2020-10-20 22:46     ` Lukasz Wojciechowski
2020-10-30 14:26     ` David Marchand
2020-10-20  0:06   ` [dpdk-dev] [PATCH 1/2] test/rcu: fix array subscript is above array bounds Lukasz Wojciechowski
2020-10-20 16:26     ` Honnappa Nagarahalli [this message]
2020-10-20 20:59       ` Lukasz Wojciechowski
2020-10-30 14:44   ` David Marchand

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=AM8PR08MB5810C4B08673FA071091719D981F0@AM8PR08MB5810.eurprd08.prod.outlook.com \
    --to=honnappa.nagarahalli@arm.com \
    --cc=Ruifeng.Wang@arm.com \
    --cc=david.marchand@redhat.com \
    --cc=dev@dpdk.org \
    --cc=john.mcnamara@intel.com \
    --cc=juraj.linkes@pantheon.tech \
    --cc=l.wojciechow@partner.samsung.com \
    --cc=nd@arm.com \
    --cc=stable@dpdk.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 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).