[PATCH v2 3/3] test/bbdev: explicit check for allocation failure

[Maxime Coquelin]

Hi Nicolas,

On 12/7/22 16:33, Chautru, Nicolas wrote:
> Hi Maxime,
> 
> I don’t believe this constitutes a fix, just adding more error check to be more comprehensive. Not required for stable branch really. Let me know if you disagree.

I think we should backport it since it constitutes a fix.
Without it, user could face a segfault, so it fixes a potential
segfault.


> Can this be applied?

Yes, with fixes tags.
I can add them while applying if you agree.

Regards,
Maxime

> Thanks
> Nic
> 
>> -----Original Message-----
>> From: Maxime Coquelin <maxime.coquelin at redhat.com>
>> Sent: Wednesday, December 7, 2022 12:40 AM
>> To: Chautru, Nicolas <nicolas.chautru at intel.com>; dev at dpdk.org
>> Cc: Vargas, Hernan <hernan.vargas at intel.com>; Vipin.Varghese at amd.com
>> Subject: Re: [PATCH v2 3/3] test/bbdev: explicit check for allocation failure
>>
>>
>>
>> On 12/6/22 20:14, Nicolas Chautru wrote:
>>> Adding check for error on return of the rte_bbdev_*_op_alloc_bulk
>>> calls in bbdev-test.
>>>
>>> Signed-off-by: Nicolas Chautru <nicolas.chautru at intel.com>
>>> ---
>>>    app/test-bbdev/test_bbdev_perf.c | 9 ++++++---
>>>    1 file changed, 6 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/app/test-bbdev/test_bbdev_perf.c
>>> b/app/test-bbdev/test_bbdev_perf.c
>>> index b2e536b5e3..a922d69ada 100644
>>> --- a/app/test-bbdev/test_bbdev_perf.c
>>> +++ b/app/test-bbdev/test_bbdev_perf.c
>>> @@ -4958,7 +4958,8 @@ offload_latency_test_fft(struct rte_mempool
>> *mempool, struct test_buffers *bufs,
>>>    		if (unlikely(num_to_process - dequeued < burst_sz))
>>>    			burst_sz = num_to_process - dequeued;
>>>
>>> -		rte_bbdev_fft_op_alloc_bulk(mempool, ops_enq, burst_sz);
>>> +		ret = rte_bbdev_fft_op_alloc_bulk(mempool, ops_enq,
>> burst_sz);
>>> +		TEST_ASSERT_SUCCESS(ret, "Allocation failed for %d ops",
>> ops_enq);
>>>    		if (test_vector.op_type != RTE_BBDEV_OP_NONE)
>>>    			copy_reference_fft_op(ops_enq, burst_sz,
>> dequeued,
>>>    					bufs->inputs,
>>> @@ -5040,7 +5041,8 @@ offload_latency_test_dec(struct rte_mempool
>> *mempool, struct test_buffers *bufs,
>>>    		if (unlikely(num_to_process - dequeued < burst_sz))
>>>    			burst_sz = num_to_process - dequeued;
>>>
>>> -		rte_bbdev_dec_op_alloc_bulk(mempool, ops_enq,
>> burst_sz);
>>> +		ret = rte_bbdev_dec_op_alloc_bulk(mempool, ops_enq,
>> burst_sz);
>>> +		TEST_ASSERT_SUCCESS(ret, "Allocation failed for %d ops",
>> ops_enq);
>>>    		if (test_vector.op_type != RTE_BBDEV_OP_NONE)
>>>    			copy_reference_dec_op(ops_enq, burst_sz,
>> dequeued,
>>>    					bufs->inputs,
>>> @@ -5125,7 +5127,8 @@ offload_latency_test_ldpc_dec(struct
>> rte_mempool *mempool,
>>>    		if (unlikely(num_to_process - dequeued < burst_sz))
>>>    			burst_sz = num_to_process - dequeued;
>>>
>>> -		rte_bbdev_dec_op_alloc_bulk(mempool, ops_enq,
>> burst_sz);
>>> +		ret = rte_bbdev_dec_op_alloc_bulk(mempool, ops_enq,
>> burst_sz);
>>> +		TEST_ASSERT_SUCCESS(ret, "Allocation failed for %d ops",
>> ops_enq);
>>>    		if (test_vector.op_type != RTE_BBDEV_OP_NONE)
>>>    			copy_reference_ldpc_dec_op(ops_enq, burst_sz,
>> dequeued,
>>>    					bufs->inputs,
>>
>> You need to add Fixes tag and cc stable.
>>
>> Maxime
>