[Environment]
DPDK version:  
commit 684d996a201182b5335012055df33284e753d832 (tag: v22.11.8-rc1,
origin/22.11)
Author: Luca Boccassi <luca.boccassi@gmail.com>
Date:   Wed Mar 26 11:26:26 2025 +0000    version: 22.11.8-rc1   
Signed-off-by: Luca Boccassi <luca.boccassi@gmail.com>
OS:  Ubuntu 18.04.5 LTS/5.4.0-150-generic
Compiler:  gcc version 7.5.0
Hardware platform:  Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz
intel-ipsec-mb: tag v1.5

[Test Setup]
Steps to reproduce
List the steps to reproduce the issue.

1, Build dpdk

rm -rf x86_64-native-linuxapp-gcc
CC=gcc meson -Dlibdir=lib  --default-library=static x86_64-native-linuxapp-gcc
ninja -C x86_64-native-linuxapp-gcc

2, Run unit tests: cryptodev_aesni_gcm_autotest cryptodev_aesni_mb_autotest
cryptodev_openssl_autotest

x86_64-native-linuxapp-gcc/app/test/dpdk-test -c 0xff --vdev=crypto_aesni_mb
--vdev=crypto_aesni_gcm --vdev=crypto_openssl
RTE>>cryptodev_aesni_gcm_autotest 

[Actual Result]:
 root@dpdk-yaobing-purely147:~/dpdk#
x86_64-native-linuxapp-gcc/app/test/dpdk-test -c 0xff --vdev=crypto_aesni_mb
--vdev=crypto_aesni_gcm --vdev=crypto_openssl -a 0000:00:00.0
EAL: Detected CPU lcores: 112
EAL: Detected NUMA nodes: 2
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
EAL: VFIO support initialized
CRYPTODEV: Creating cryptodev crypto_aesni_mb
CRYPTODEV: Initialisation parameters - name: crypto_aesni_mb,socket id: 0, max
queue pairs: 8
ipsec_mb_create() line 169: IPSec Multi-buffer library version used:
1.5.0CRYPTODEV: Creating cryptodev crypto_aesni_gcm
CRYPTODEV: Initialisation parameters - name: crypto_aesni_gcm,socket id: 0, max
queue pairs: 8
ipsec_mb_create() line 169: IPSec Multi-buffer library version used:
1.5.0CRYPTODEV: Creating cryptodev crypto_openssl
CRYPTODEV: Initialisation parameters - name: crypto_openssl,socket id: 0, max
queue pairs: 8
APP: HPET is not enabled, using TSC as default timer
RTE>>cryptodev_aesni_gcm_autotest
 + TestCase [55] : test_AES_GCM_authenticated_decryption_oop_test_case_1
succeeded
aad: at [0x17d9e3700], len=8
00000000: FE ED FA CE DE AD BE EF                         | ........
iv: at [0x17d5275f8], len=12
00000000: CA FE BA BE FA CE DB AD DE CA F8 88             | ............
plaintext: at [0x17d9e3710], len=60
00000000: D9 31 32 25 F8 84 06 E5 A5 59 09 C5 AF F5 26 9A | .12%.....Y....&.
00000010: 86 A7 A9 53 15 34 F7 DA 2E 4C 30 3D 8A 31 8A 72 | ...S.4...L0=.1.r
00000020: 1C 3C 0C 95 95 68 09 53 2F CF 0E 24 49 A6 B5 25 | .<...h.S/..$I..%
00000030: B1 6A ED F5 AA 0D E6 57 BA 63 7B 39             | .j.....W.c{9
key: at [0x7ffe52c4eb20], len=16
00000000: FE FF E9 92 86 65 73 1C 6D 6A 8F 94 67 30 83 08 | .....es.mj..g0..
Segmentation fault (core dumped)
root@dpdk-yaobing-purely147:~/dpdk#

[Expected Result]
Test OK

Is this issue a regression: Y
Version the regression was introduced: 
Bad commit bb8feaacb4897e3ee4341dd838de538869d8225e (HEAD -> bb8feaacb4)
Author: Arkadiusz Kusztal <arkadiuszx.kusztal@intel.com>
Date:   Wed Feb 26 08:16:45 2025 +0000

    test/crypto: fix check for OOP header data

    [ upstream commit 27eb74ea6e25966d3857539cb15ddd4f20f05ebe ]

    The data of the out-of-place header is never checked. Therefore,
    the faulty PMD, which overwrites this data, will not be able to
    verify that with tests. New checks to support that were added to
    the GCM OOP functions.

    Fixes: 51e202f0596f ("test/crypto: rename GCM test code")

    Signed-off-by: Arkadiusz Kusztal <arkadiuszx.kusztal@intel.com>

Stack Trace or Log
1, Revert the bad commit, unit test can work well.
2, Impact three cases:　
    cryptodev_aesni_gcm_autotest 
    cryptodev_aesni_mb_autotest 
    cryptodev_openssl_autotest