denozord/strongswan-ext
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
18,891 Commits 1 Branch 0 Tags
67f09905305c93620f552b8f9084b5df40e920cf
Commit Graph
32 Commits
Author SHA1 Message Date
Tobias Brunner ad8484a6cc aesni: Make sure to wipe salt 2023-07-26 15:08:33 +02:00
Tobias Brunner 7f46c76125 Make functions static that are only accessed from the same compilation unit 
Also removed some declarations for undefined functions.
 2022-10-05 19:00:46 +02:00
Tobias Brunner 7217ff5fc5 aesni: Fix out-of-bound read when loading 192-bit AES keys 2022-09-15 12:16:12 +02:00
Tobias Brunner 19ef2aec15 Update copyright headers after acquisition by secunet 2022-06-28 10:22:56 +02:00
Tobias Brunner 3f8eb2ebdf aesni: Remove useless algorithm assignments 2020-08-18 13:17:24 +02:00
Tobias Brunner 26f20cc258 aesni: Namespace include guard for AES-CMAC 
Was the same as in the cmac plugin.
 2020-01-28 15:32:43 +01:00
Andreas Steffen 20f3d04b13 aesni: Added AES_ECB support 2019-11-28 17:03:08 +01:00
Tobias Brunner 1b67166921 Unify format of HSR copyright statements 2018-05-23 16:32:53 +02:00
Andreas Steffen b12c53ce77 Use standard unsigned integer types 2016-03-24 18:52:48 +01:00
Martin Willi 23947b2a4f aesni: Fix doxygen groups 2015-04-15 17:29:56 +02:00
Martin Willi 37794878cc aesni: Avoid loading AES/GHASH round keys into local variables 
The performance impact is not measurable, as the compiler loads these variables
in xmm registers in unrolled loops anyway.

However, we avoid loading these sensitive keys onto the stack. This happens for
larger key schedules, where the register count is insufficient. If that key
material is not on the stack, we can avoid to wipe it explicitly after
crypto operations.
 2015-04-15 13:44:40 +02:00
Martin Willi 93f0080265 aesni: Align all class instances to 16 byte boundaries 
While the required members are aligned in the struct as required, on 32-bit
platforms the allocator aligns the structures itself to 8 bytes only. This
results in non-aligned struct members, and invalid memory accesses.
 2015-04-15 13:44:40 +02:00
Martin Willi edab6c658c aesni: Calculate GHASH for 4 blocks of associated data in parallel 
While associated data is usually not that large, in some specific cases
this can bring a significant performance boost.
 2015-04-15 11:35:28 +02:00
Martin Willi 0eb593b0bb aesni: Calculate GHASH for 4 blocks of encryption data in parallel 
Increases performance by another ~30%.
 2015-04-15 11:35:28 +02:00
Martin Willi 58c44cdd00 aesni: Use 4-way parallel en/decryption in GCM 
Increases overall performance by ~25%.
 2015-04-15 11:35:28 +02:00
Martin Willi 677649cfb5 aesni: Use dedicated key size specific en-/decryption functions in GCM 
This gives not much more than ~5% increase in performance, but allows us to
improve further.
 2015-04-15 11:35:28 +02:00
Martin Willi 313811b72d aesni: Add a GCM AEAD based on the AES-NI key schedule 2015-04-15 11:35:28 +02:00
Martin Willi 4284660677 aesni: Implement CMAC mode to provide a signer/prf 
Compared to the cmac plugin using AESNI-CBC as backend, this improves
performance of AES-CMAC by ~45%.
 2015-04-15 11:35:28 +02:00
Martin Willi 9d187f8711 aesni: Implement XCBC mode to provide a signer/prf 
Compared to the xcbc plugin using AESNI-CBC as backend, this improves
performance of AES-XCBC by ~45%.
 2015-04-15 11:35:28 +02:00
Martin Willi d48642b7eb aesni: Partially use separate code paths for different key sizes in CCM 
Due to the serial nature of the CBC mac, this brings only a marginal speedup.
 2015-04-15 11:35:28 +02:00
Martin Willi 1ae46dfba2 aesni: Add a CCM AEAD reusing the key schedule 2015-04-15 11:35:27 +02:00
Martin Willi 8488dea2f9 aesni: Use 4-way parallel AES-NI instructions for CTR en/decryption 
CTR can be parallelized, and we do so by queueing instructions to the processor
pipeline. While we have enough registers for 128-bit decryption, the register
count is insufficient to hold all variables with larger key sizes. Nonetheless
is 4-way parallelism faster, depending on key size between ~10% and ~25%.
 2015-04-15 11:35:27 +02:00
Martin Willi 9e47c1fe77 aesni: Use dedicated round count specific encryption functions in CTR mode 
This allows us to unroll loops and hold the key schedule in local (register)
variables. This brings an impressive speedup of ~45%.
 2015-04-15 11:35:27 +02:00
Martin Willi 74d43cbde9 aesni: Implement a AES-NI based CTR crypter using the key schedule 2015-04-15 11:35:27 +02:00
Martin Willi a3cb72e850 aesni: Use 4-way parallel AES-NI instructions for CBC decryption 
CBC decryption can be parallelized, and we do so by queueing instructions
to the processor pipeline. While we have enough registers for 128-bit
decryption, the register count is insufficient to hold all variables with
larger key sizes. Nonetheless is 4-way parallelism faster, roughly by ~8%.
 2015-04-15 11:35:27 +02:00
Martin Willi 95ef361415 aesni: Use separate en-/decryption CBC code paths for different key sizes 
This allows us to unroll loops, and use local (register) variables for the
key schedule. This improves performance slightly for encryption, but a lot
for reorderable decryption (>30%).
 2015-04-15 11:35:27 +02:00
Martin Willi 8f1fab93e9 aesni: Implement a AES-NI based CBC crypter using the key schedule 2015-04-15 11:35:27 +02:00
Martin Willi b2d96bfe25 aesni: Implement 256-bit key schedule 2015-04-15 11:35:27 +02:00
Martin Willi d7811415ff aesni: Implement 192-bit key schedule 2015-04-15 11:35:26 +02:00
Martin Willi 6d7671eb1c aesni: Implement 128-bit key schedule 2015-04-15 11:35:26 +02:00
Martin Willi 431f452ed4 aesni: Add a common key schedule class for AES 2015-04-15 11:35:26 +02:00
Martin Willi 78c04b5d4d aesni: Provide a plugin stub for AES-NI instruction based crypto primitives 2015-04-15 11:35:26 +02:00