- 11 Nov, 2022 3 commits
-
-
Ard Biesheuvel authored
Implement a minimal library version of AES-GCM based on the existing library implementations of AES and multiplication in GF(2^128). Using these primitives, GCM can be implemented in a straight-forward manner. GCM has a couple of sharp edges, i.e., the amount of input data processed with the same initialization vector (IV) should be capped to protect the counter from 32-bit rollover (or carry), and the size of the authentication tag should be fixed for a given key. [0] The former concern is addressed trivially, given that the function call API uses 32-bit signed types for the input lengths. It is still up to the caller to avoid IV reuse in general, but this is not something we can police at the implementation level. As for the latter concern, let's make the authentication tag size part of the key schedule, and only permit it to be configured as part of the key expansion routine. Note that table based AES implementations are susceptible to known plaintext timing attacks on the encryption key. The AES library already attempts to mitigate this to some extent, but given that the counter mode encryption used by GCM operates exclusively on known plaintext by construction (the IV and therefore the initial counter value are known to an attacker), let's take some extra care to mitigate this, by calling the AES library with interrupts disabled. [0] https://nvlpubs.nist.gov/nistpubs/legacy/sp/nistspecialpublication800-38d.pdf Link: https://lore.kernel.org/all/c6fb9b25-a4b6-2e4a-2dd1-63adda055a49@amd.com/Signed-off-by:
Ard Biesheuvel <ardb@kernel.org> Tested-by:
Nikunj A Dadhania <nikunj@amd.com> Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au>
-
Ard Biesheuvel authored
The gf128mul library has different variants with different memory/performance tradeoffs, where the faster ones use 4k or 64k lookup tables precomputed at runtime, which are based on one of the multiplication factors, which is commonly the key for keyed hash algorithms such as GHASH. The slowest variant is gf128_mul_lle() [and its bbe/ble counterparts], which does not use precomputed lookup tables, but it still relies on a single u16[256] lookup table which is input independent. The use of such a table may cause the execution time of gf128_mul_lle() to correlate with the value of the inputs, which is generally something that must be avoided for cryptographic algorithms. On top of that, the function uses a sequence of if () statements that conditionally invoke be128_xor() based on which bits are set in the second argument of the function, which is usually a pointer to the multiplication factor that represents the key. In order to remove the correlation between the execution time of gf128_mul_lle() and the value of its inputs, let's address the identified shortcomings: - add a time invariant version of gf128mul_x8_lle() that replaces the table lookup with the expression that is used at compile time to populate the lookup table; - make the invocations of be128_xor() unconditional, but pass a zero vector as the third argument if the associated bit in the key is cleared. The resulting code is likely to be significantly slower. However, given that this is the slowest version already, making it even slower in order to make it more secure is assumed to be justified. The bbe and ble counterparts could receive the same treatment, but the former is never used anywhere in the kernel, and the latter is only used in the driver for a asynchronous crypto h/w accelerator (Chelsio), where timing variances are unlikely to matter. Signed-off-by:
-
Ard Biesheuvel authored
The gf128mul library does not depend on the crypto API at all, so it can be moved into lib/crypto. This will allow us to use it in other library code in a subsequent patch without having to depend on CONFIG_CRYPTO. While at it, change the Kconfig symbol name to align with other crypto library implementations. However, the source file name is retained, as it is reflected in the module .ko filename, and changing this might break things for users. Signed-off-by:
-
- 04 Nov, 2022 18 commits
-
-
Ralph Siemsen authored
The hashing API does not have a function called .finish() Signed-off-by:
Ralph Siemsen <ralph.siemsen@linaro.org> Signed-off-by:
-
Tianjia Zhang authored
This patch is a CE-optimized assembly implementation for GCM mode. Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 224 and 224 modes of tcrypt, and compared the performance before and after this patch (the driver used before this patch is gcm_base(ctr-sm4-ce,ghash-generic)). The abscissas are blocks of different lengths. The data is tabulated and the unit is Mb/s: Before (gcm_base(ctr-sm4-ce,ghash-generic)): gcm(sm4) | 16 64 256 512 1024 1420 4096 8192 -------------+--------------------------------------------------------------------- GCM enc | 25.24 64.65 104.66 116.69 123.81 125.12 129.67 130.62 GCM dec | 25.40 64.80 104.74 116.70 123.81 125.21 129.68 130.59 GCM mb enc | 24.95 64.06 104.20 116.38 123.55 124.97 129.63 130.61 GCM mb dec | 24.92 64.00 104.13 116.34 123.55 124.98 129.56 130.48 After: gcm-sm4-ce | 16 64 256 512 1024 1420 4096 8192 -------------+--------------------------------------------------------------------- GCM enc | 108.62 397.18 971.60 1283.92 1522.77 1513.39 1777.00 1806.96 GCM dec | 116.36 398.14 1004.27 1319.11 1624.21 1635.43 1932.54 1974.20 GCM mb enc | 107.13 391.79 962.05 1274.94 1514.76 1508.57 1769.07 1801.58 GCM mb dec | 113.40 389.36 988.51 1307.68 1619.10 1631.55 1931.70 1970.86 Signed-off-by:
Tianjia Zhang <tianjia.zhang@linux.alibaba.com> Signed-off-by:
-
Tianjia Zhang authored
This patch is a CE-optimized assembly implementation for CCM mode. Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 223 and 225 modes of tcrypt, and compared the performance before and after this patch (the driver used before this patch is ccm_base(ctr-sm4-ce,cbcmac-sm4-ce)). The abscissas are blocks of different lengths. The data is tabulated and the unit is Mb/s: Before (rfc4309(ccm_base(ctr-sm4-ce,cbcmac-sm4-ce))): ccm(sm4) | 16 64 256 512 1024 1420 4096 8192 -------------+--------------------------------------------------------------- CCM enc | 35.07 125.40 336.47 468.17 581.97 619.18 712.56 736.01 CCM dec | 34.87 124.40 335.08 466.75 581.04 618.81 712.25 735.89 CCM mb enc | 34.71 123.96 333.92 465.39 579.91 617.49 711.45 734.92 CCM mb dec | 34.42 122.80 331.02 462.81 578.28 616.42 709.88 734.19 After (rfc4309(ccm-sm4-ce)): ccm-sm4-ce | 16 64 256 512 1024 1420 4096 8192 -------------+--------------------------------------------------------------- CCM enc | 77.12 249.82 569.94 725.17 839.27 867.71 952.87 969.89 CCM dec | 75.90 247.26 566.29 722.12 836.90 865.95 951.74 968.57 CCM mb enc | 75.98 245.25 562.91 718.99 834.76 864.70 950.17 967.90 CCM mb dec | 75.06 243.78 560.58 717.13 833.68 862.70 949.35 967.11 Signed-off-by:
-
Tianjia Zhang authored
This patch is a CE-optimized assembly implementation for cmac/xcbc/cbcmac. Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 300 mode of tcrypt, and compared the performance before and after this patch (the driver used before this patch is XXXmac(sm4-ce)). The abscissas are blocks of different lengths. The data is tabulated and the unit is Mb/s: Before: update-size | 16 64 256 1024 2048 4096 8192 ---------------+-------------------------------------------------------- cmac(sm4-ce) | 293.33 403.69 503.76 527.78 531.10 535.46 535.81 xcbc(sm4-ce) | 292.83 402.50 504.02 529.08 529.87 536.55 538.24 cbcmac(sm4-ce) | 318.42 415.79 497.12 515.05 523.15 521.19 523.01 After: update-size | 16 64 256 1024 2048 4096 8192 ---------------+-------------------------------------------------------- cmac-sm4-ce | 371.99 675.28 903.56 971.65 980.57 990.40 991.04 xcbc-sm4-ce | 372.11 674.55 903.47 971.61 980.96 990.42 991.10 cbcmac-sm4-ce | 371.63 675.33 903.23 972.07 981.42 990.93 991.45 Signed-off-by:
-
Tianjia Zhang authored
This patch is a CE-optimized assembly implementation for XTS mode. Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 218 mode of tcrypt, and compared the performance before and after this patch (the driver used before this patch is xts(ecb-sm4-ce)). The abscissas are blocks of different lengths. The data is tabulated and the unit is Mb/s: Before: xts(ecb-sm4-ce) | 16 64 128 256 1024 1420 4096 ----------------+-------------------------------------------------------------- XTS enc | 117.17 430.56 732.92 1134.98 2007.03 2136.23 2347.20 XTS dec | 116.89 429.02 733.40 1132.96 2006.13 2130.50 2347.92 After: xts-sm4-ce | 16 64 128 256 1024 1420 4096 ----------------+-------------------------------------------------------------- XTS enc | 224.68 798.91 1248.08 1714.60 2413.73 2467.84 2612.62 XTS dec | 229.85 791.34 1237.79 1720.00 2413.30 2473.84 2611.95 Signed-off-by: -
Tianjia Zhang authored
This patch is a CE-optimized assembly implementation for CTS-CBC mode. Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 218 mode of tcrypt, and compared the performance before and after this patch (the driver used before this patch is cts(cbc-sm4-ce)). The abscissas are blocks of different lengths. The data is tabulated and the unit is Mb/s: Before: cts(cbc-sm4-ce) | 16 64 128 256 1024 1420 4096 ----------------+-------------------------------------------------------------- CTS-CBC enc | 286.09 297.17 457.97 627.75 868.58 900.80 957.69 CTS-CBC dec | 286.67 285.63 538.35 947.08 2241.03 2577.32 3391.14 After: cts-cbc-sm4-ce | 16 64 128 256 1024 1420 4096 ----------------+-------------------------------------------------------------- CTS-CBC enc | 288.19 428.80 593.57 741.04 911.73 931.80 950.00 CTS-CBC dec | 292.22 468.99 838.23 1380.76 2741.17 3036.42 3409.62 Signed-off-by: -
Tianjia Zhang authored
In the accelerated implementation of the SM4 algorithm using the Crypto Extension instructions, there are some functions that can be reused in the upcoming accelerated implementation of the GCM/CCM mode, and the CBC/CFB encryption is reused in the optimized implementation of SVESM4. Signed-off-by:
-
Tianjia Zhang authored
Use a 128-bit swap mask and tbl instruction to simplify the implementation for generating SM4 rkey_dec. Also fixed the issue of not being wrapped by kernel_neon_begin/end() when using the sm4_ce_expand_key() function. Signed-off-by:
-
Tianjia Zhang authored
This patch does not add new features, but only refactors and simplifies the implementation of the Crypto Extension acceleration of the SM4 algorithm: Extract the macro optimized by SM4 Crypto Extension for reuse in the subsequent optimization of CCM/GCM modes. Encryption in CBC and CFB modes processes four blocks at a time instead of one, allowing the ld1 instruction to load 64 bytes of data at a time, which will reduces unnecessary memory accesses. CBC/CFB/CTR makes full use of free registers to reduce redundant memory accesses, and rearranges some instructions to improve out-of-order execution capabilities. Signed-off-by:
-
Tianjia Zhang authored
Added CTS-CBC/XTS/XCBC tests for SM4 algorithms, as well as corresponding speed tests, this is to test performance-optimized implementations of these modes. Signed-off-by:
-
Tianjia Zhang authored
This patch newly adds the test vectors of CTS-CBC/XTS/XCBC modes of the SM4 algorithm, and also added some test vectors for SM4 GCM/CCM. Signed-off-by:
-
Tianjia Zhang authored
This patch does not add new features. The main work is to refactor and simplify the implementation of SM4 NEON, which is reflected in the following aspects: The accelerated implementation supports the arbitrary number of blocks, not just multiples of 8, which simplifies the implementation and brings some optimization acceleration for data that is not aligned by 8 blocks. When loading the input data, use the ld4 instruction to replace the original ld1 instruction as much as possible, which will save the cost of matrix transposition of the input data. Use 8-block parallelism whenever possible to speed up matrix transpose and rotation operations, instead of up to 4-block parallelism. Signed-off-by:
-
Tianjia Zhang authored
This patch adds the NEON acceleration implementation of the SM3 hash algorithm. The main algorithm is based on SM3 NEON accelerated work of the libgcrypt project. Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 326 mode of tcrypt, and compares the performance data of sm3-generic and sm3-ce. The abscissas are blocks of different lengths. The data is tabulated and the unit is Mb/s: update-size | 16 64 256 1024 2048 4096 8192 ---------------+-------------------------------------------------------- sm3-generic | 185.24 221.28 301.26 307.43 300.83 308.82 308.91 sm3-neon | 171.81 220.20 322.94 339.28 334.09 343.61 343.87 sm3-ce | 227.48 333.48 502.62 527.87 520.45 534.91 535.40 Signed-off-by:
-
Tianjia Zhang authored
Raise the priority of the sm3-ce algorithm from 200 to 400, this is to make room for the implementation of sm3-neon. Signed-off-by:
-
Anirudh Venkataramanan authored
The top level print banners have a leading newline. It's not entirely clear why this exists, but it makes it harder to parse tcrypt test output using a script. Drop said newlines. tcrypt output before this patch: [...] testing speed of rfc4106(gcm(aes)) (rfc4106-gcm-aesni) encryption [...] test 0 (160 bit key, 16 byte blocks): 1 operation in 2320 cycles (16 bytes) tcrypt output with this patch: [...] testing speed of rfc4106(gcm(aes)) (rfc4106-gcm-aesni) encryption [...] test 0 (160 bit key, 16 byte blocks): 1 operation in 2320 cycles (16 bytes) Signed-off-by:Anirudh Venkataramanan <anirudh.venkataramanan@intel.com> Signed-off-by:
-
Anirudh Venkataramanan authored
The pr_fmt() define includes KBUILD_MODNAME, and so there's no need for pr_err() to also print it. Drop module name from the print string. Signed-off-by:
-
Anirudh Venkataramanan authored
Currently, there's mixed use of printk() and pr_info()/pr_err(). The latter prints the module name (because pr_fmt() is defined so) but the former does not. As a result there's inconsistency in the printed output. For example: modprobe mode=211: [...] test 0 (160 bit key, 16 byte blocks): 1 operation in 2320 cycles (16 bytes) [...] test 1 (160 bit key, 64 byte blocks): 1 operation in 2336 cycles (64 bytes) modprobe mode=215: [...] tcrypt: test 0 (160 bit key, 16 byte blocks): 1 operation in 2173 cycles (16 bytes) [...] tcrypt: test 1 (160 bit key, 64 byte blocks): 1 operation in 2241 cycles (64 bytes) Replace all instances of printk() with pr_info()/pr_err() so that the module name is printed consistently. Signed-off-by:
-
Anirudh Venkataramanan authored
For some test cases, a line break gets inserted between the test banner and the results. For example, with mode=211 this is the output: [...] testing speed of rfc4106(gcm(aes)) (rfc4106-gcm-aesni) encryption [...] test 0 (160 bit key, 16 byte blocks): [...] 1 operation in 2373 cycles (16 bytes) --snip-- [...] testing speed of gcm(aes) (generic-gcm-aesni) encryption [...] test 0 (128 bit key, 16 byte blocks): [...] 1 operation in 2338 cycles (16 bytes) Similar behavior is seen in the following cases as well: modprobe tcrypt mode=212 modprobe tcrypt mode=213 modprobe tcrypt mode=221 modprobe tcrypt mode=300 sec=1 modprobe tcrypt mode=400 sec=1 This doesn't happen with mode=215: [...] tcrypt: testing speed of multibuffer rfc4106(gcm(aes)) (rfc4106-gcm-aesni) encryption [...] tcrypt: test 0 (160 bit key, 16 byte blocks): 1 operation in 2215 cycles (16 bytes) --snip-- [...] tcrypt: testing speed of multibuffer gcm(aes) (generic-gcm-aesni) encryption [...] tcrypt: test 0 (128 bit key, 16 byte blocks): 1 operation in 2191 cycles (16 bytes) This print inconsistency is because printk() is used instead of pr_cont() in a few places. Change these to be pr_cont(). checkpatch warns that pr_cont() shouldn't be used. This can be ignored in this context as tcrypt already uses pr_cont(). Signed-off-by:
-
- 28 Oct, 2022 19 commits
-
-
wangjianli authored
Delete the redundant word 'the'. Signed-off-by:
wangjianli <wangjianli@cdjrlc.com> Signed-off-by:
-
Kai Ye authored
Because the permission on the VF debugfs file is "0444". So the VF function checking is redundant in qos writing api. Signed-off-by:
Kai Ye <yekai13@huawei.com> Signed-off-by:
-
Kai Ye authored
The pci bdf number check is added for qos written by using the pci api. Directly get the devfn by pci_dev, so delete some redundant code. And use the kstrtoul instead of sscanf to simplify code. Signed-off-by:
-
Kai Ye authored
Increase the buffer to prevent stack overflow by fuzz test. The maximum length of the qos configuration buffer is 256 bytes. Currently, the value of the 'val buffer' is only 32 bytes. The sscanf does not check the dest memory length. So the 'val buffer' may stack overflow. Signed-off-by:
-
Frederick Lawler authored
We want to leverage keyring to store sensitive keys, and then use those keys for symmetric encryption via the crypto API. Among the key types we wish to support are: user, logon, encrypted, and trusted. User key types are already able to have their data copied to user space, but logon does not support this. Further, trusted and encrypted keys will return their encrypted data back to user space on read, which does not make them ideal for symmetric encryption. To support symmetric encryption for these key types, add a new ALG_SET_KEY_BY_KEY_SERIAL setsockopt() option to the crypto API. This allows users to pass a key_serial_t to the crypto API to perform symmetric encryption. The behavior is the same as ALG_SET_KEY, but the crypto key data is copied in kernel space from a keyring key, which allows for the support of logon, encrypted, and trusted key types. Keyring keys must have the KEY_(POS|USR|GRP|OTH)_SEARCH permission set to leverage this feature. This follows the asymmetric_key type where key lookup calls eventually lead to keyring_search_rcu() without the KEYRING_SEARCH_NO_CHECK_PERM flag set. Signed-off-by:
Frederick Lawler <fred@cloudflare.com> Signed-off-by:
-
Corentin Labbe authored
The RK3399 has 2 rk3288 compatible crypto device named crypto0 and crypto1. The only difference is lack of RSA in crypto1. We need to add driver support for 2 parallel instance as only one need to register crypto algorithms. Then the driver will round robin each request on each device. For avoiding complexity (device bringup after a TFM is created), PM is modified to be handled per request. Signed-off-by:
Corentin Labbe <clabbe@baylibre.com> Signed-off-by:
-
Corentin Labbe authored
The RK3399 has 3 resets, so the driver to handle multiple resets. This is done by using devm_reset_control_array_get_exclusive(). Signed-off-by:
-
Corentin Labbe authored
rk_ahash_reg_init() use crypto_info from TFM context, since we will remove it, let's take if from parameters. Signed-off-by:
-
Corentin Labbe authored
Add the number of clocks needed for each compatible. Rockchip's datasheet give maximum frequencies for some clocks, so add checks for verifying they are within limits. Let's start with rk3288 for clock frequency check, other will came later. Signed-off-by:
-
Corentin Labbe authored
The crypto_info to use must be stored in the request context. This will help when 2 crypto_info will be available on rk3399. Signed-off-by:
-
Corentin Labbe authored
Since driver support new compatible, we need to update the driver bindings. Signed-off-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Corentin Labbe authored
Convert rockchip-crypto to YAML. Reviewed-by:
John Keeping <john@metanate.com> Reviewed-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by:
-
Corentin Labbe authored
Instead of using the crypto_info from TFM ctx, use the one given as parameter. Reviewed-by:
-
Corentin Labbe authored
Instead of using lot of ctx->dev->xx indirections, use an intermediate variable for rk_crypto_info. This will help later, when 2 different rk_crypto_info would be used. Reviewed-by:
-
Corentin Labbe authored
This patch rework the rk_handle_req(), simply removing the rk_crypto_info parameter. Reviewed-by:
-
Corentin Labbe authored
Some functions have still ablk in their name even if there are not handling ablk_cipher anymore. So let's rename them. Reviewed-by:
-
Corentin Labbe authored
The rk3328 could be used as-is by the rockchip driver. Reviewed-by:
-
Corentin Labbe authored
This patch fixes some warning reported by checkpatch Reviewed-by:
-
Corentin Labbe authored
Use read_poll_timeout instead of open coding it. In the same time, fix indentation of related comment. Reviewed-by:
-
