Hi, Noticed this issue in linux-2.6.34.14 and its same in linux-3.14.4. In ablk_decrypt, we check may_use_simd() function to decrypt the packet. If may_use_simd() function returns true, we get the packet decrypted immediately and return. If it return false packet will be queued for decryption.
When queue already have packets for decryption, if for a incoming packet "may_use_simd()" function returns true, the packet will get decrypted immediately and returned with out checking the packets in queue. So the packets in the queue will be send to upper layer or forwarded late, resulting in the reordering of the packets. Is this fine? my understanding was packets will be never be reorder in the ipsec layer in kernel. Let me know if question is not clear, i will try to frame it better. Please include my email id in the response -- Guruswamy Basavaiah -- To unsubscribe from this list: send the line "unsubscribe linux-crypto" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
