>
>> The testsuite_setup function in the compressdev_autotest test case,
>> the rte_compressdev_count() return 0 due to the lack of device init.
>> It is considered that there is no equitment to skip execution.
>> So add rte_vdev_init init compress_zlib device.
>
>Why? Devices can be added as command line argument.
>* Use ``--vdev="compress_isal"`` in the EAL options, which will call
>``rte_vdev_init()`` internally.
>See documentation of compressdevs.
>
>And if at all it is required, why only the zlib, there are other devices as
>well.
>
When ut test in this steps, will not call rte_vdev_init()
1. meson build
2. cd build
3. meson test compressdev_autotest
This use case tests the framework code and doesn't care about the specific
device, so zlib device is randomly selected.
According to what you said, it may affect the command-line test. I need to
think it again.
>> Signed-off-by: Changsheng Wu <wuchangshe...@huawei.com>
>> ---
>> app/test/test_compressdev.c | 6 ++++++
>> 1 file changed, 6 insertions(+)
>>
>> diff --git a/app/test/test_compressdev.c b/app/test/test_compressdev.c
>> index a1b9f06250..f15fbe7f2d 100644
>> --- a/app/test/test_compressdev.c
>> +++ b/app/test/test_compressdev.c
>> @@ -14,6 +14,7 @@
>> #include <rte_mbuf.h>
>> #include <rte_compressdev.h>
>> #include <rte_string_fns.h>
>> +#include <rte_bus_vdev.h>
>>
>> #include "test_compressdev_test_buffer.h"
>> #include "test.h"
>> @@ -197,6 +198,11 @@ testsuite_setup(void)
>> uint32_t max_buf_size = 0;
>> unsigned int i;
>>
>> + if (rte_vdev_init("compress_zlib", NULL) != 0) {
>> + RTE_LOG(WARNING, USER1, "not find compress_zlib
>> device\n");
>> + return TEST_SKIPPED;
>> + }
>> +
>> if (rte_compressdev_count() == 0) {
>> RTE_LOG(WARNING, USER1, "Need at least one compress device\n");
>> return TEST_SKIPPED;
>> --
>> 2.27.0
