[dpdk-dev] [PATCH v2 1/2] crypto/aesni_mb: allow device init if no AES-NI is present

[Pablo de Lara]

The IPSec Multi buffer library does not require AES-NI
instructions to be supported by the CPU, as it can emulate these
instructions in software (adding a big performance penalty when
using AES algorithms).

Signed-off-by: Pablo de Lara <pablo.de.lara.gua...@intel.com>
---
v2:
- Fixed leading whitespaces

 drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c 
b/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
index ce1144b..c3001cb 100644
--- a/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
+++ b/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
@@ -1251,12 +1251,6 @@ cryptodev_aesni_mb_create(const char *name,
        enum aesni_mb_vector_mode vector_mode;
        MB_MGR *mb_mgr;
 
-       /* Check CPU for support for AES instruction set */
-       if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_AES)) {
-               AESNI_MB_LOG(ERR, "AES instructions not supported by CPU");
-               return -EFAULT;
-       }
-
        dev = rte_cryptodev_pmd_create(name, &vdev->device, init_params);
        if (dev == NULL) {
                AESNI_MB_LOG(ERR, "failed to create cryptodev vdev");
@@ -1282,9 +1276,13 @@ cryptodev_aesni_mb_create(const char *name,
 
        dev->feature_flags = RTE_CRYPTODEV_FF_SYMMETRIC_CRYPTO |
                        RTE_CRYPTODEV_FF_SYM_OPERATION_CHAINING |
-                       RTE_CRYPTODEV_FF_CPU_AESNI |
                        RTE_CRYPTODEV_FF_OOP_LB_IN_LB_OUT;
 
+       /* Check CPU for support for AES instruction set */
+       if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AES))
+               dev->feature_flags |= RTE_CRYPTODEV_FF_CPU_AESNI;
+       else
+               AESNI_MB_LOG(WARNING, "AES instructions not supported by CPU");
 
        mb_mgr = alloc_mb_mgr(0);
        if (mb_mgr == NULL)
-- 
2.7.5