Message ID | 20240305112831.3380896-1-pavitrakumarm@vayavyalabs.com (mailing list archive) |
---|---|
Headers | show |
Series | Add spacc crypto driver support | expand |
On Tue, Mar 05, 2024 at 04:58:27PM +0530, Pavitrakumar M wrote: > Add the driver for SPAcc(Security Protocol Accelerator), which is a > crypto acceleration IP from Synopsys. The SPAcc supports many cipher, > hash, aead algorithms and various modes.The driver currently supports > below, Has this been tested with CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y? > aead: > - ccm(sm4) > - ccm(aes) > - gcm(sm4) > - gcm(aes) > - rfc8998(gcm(sm4)) > - rfc7539(chacha20,poly1305) > > cipher: > - cbc(sm4) > - ecb(sm4) > - ofb(sm4) > - cfb(sm4) > - ctr(sm4) > - cbc(aes) > - ecb(aes) > - ctr(aes) > - xts(aes) > - cts(cbc(aes)) > - cbc(des) > - ecb(des) > - cbc(des3_ede) > - ecb(des3_ede) > - chacha20 > - xts(sm4) > - cts(cbc(sm4)) > - ecb(kasumi) > - f8(kasumi) > - snow3g_uea2 > - cs1(cbc(aes)) > - cs2(cbc(aes)) > - cs1(cbc(sm4)) > - cs2(cbc(sm4)) > - f8(sm4) > > hash: > - michael_mic > - sm3 > - hmac(sm3) > - sha3-512 > - sha3-384 > - sha3-256 > - sha3-224 > - hmac(sha512) > - hmac(sha384) > - hmac(sha256) > - hmac(sha224) > - sha512 > - sha384 > - sha256 > - sha224 > - sha1 > - hmac(sha1) > - md5 > - hmac(md5) > - cmac(sm4) > - xcbc(aes) > - cmac(aes) > - xcbc(sm4) > - sha512-224 > - hmac(sha512-224) > - sha512-256 > - hmac(sha512-256) > - mac(kasumi_f9) > - mac(snow3g) > - mac(zuc) > - sslmac(sha1) > - shake128 > - shake256 > - cshake128 > - cshake256 > - kcmac128 > - kcmac256 > - kcmacxof128 > - kcmacxof256 > - sslmac(md5) Algorithms that don't have a generic implementation shouldn't be included. - Eric
Hi Eric, Yes we have tested this with CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y Also is it fine if those additional algos stay inside that disabled CONFIG. SPAcc hardware does support those algos. - PK On Wed, Mar 6, 2024 at 2:43 AM Eric Biggers <ebiggers@kernel.org> wrote: > > On Tue, Mar 05, 2024 at 04:58:27PM +0530, Pavitrakumar M wrote: > > Add the driver for SPAcc(Security Protocol Accelerator), which is a > > crypto acceleration IP from Synopsys. The SPAcc supports many cipher, > > hash, aead algorithms and various modes.The driver currently supports > > below, > > Has this been tested with CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y? > > > aead: > > - ccm(sm4) > > - ccm(aes) > > - gcm(sm4) > > - gcm(aes) > > - rfc8998(gcm(sm4)) > > - rfc7539(chacha20,poly1305) > > > > cipher: > > - cbc(sm4) > > - ecb(sm4) > > - ofb(sm4) > > - cfb(sm4) > > - ctr(sm4) > > - cbc(aes) > > - ecb(aes) > > - ctr(aes) > > - xts(aes) > > - cts(cbc(aes)) > > - cbc(des) > > - ecb(des) > > - cbc(des3_ede) > > - ecb(des3_ede) > > - chacha20 > > - xts(sm4) > > - cts(cbc(sm4)) > > - ecb(kasumi) > > - f8(kasumi) > > - snow3g_uea2 > > - cs1(cbc(aes)) > > - cs2(cbc(aes)) > > - cs1(cbc(sm4)) > > - cs2(cbc(sm4)) > > - f8(sm4) > > > > hash: > > - michael_mic > > - sm3 > > - hmac(sm3) > > - sha3-512 > > - sha3-384 > > - sha3-256 > > - sha3-224 > > - hmac(sha512) > > - hmac(sha384) > > - hmac(sha256) > > - hmac(sha224) > > - sha512 > > - sha384 > > - sha256 > > - sha224 > > - sha1 > > - hmac(sha1) > > - md5 > > - hmac(md5) > > - cmac(sm4) > > - xcbc(aes) > > - cmac(aes) > > - xcbc(sm4) > > - sha512-224 > > - hmac(sha512-224) > > - sha512-256 > > - hmac(sha512-256) > > - mac(kasumi_f9) > > - mac(snow3g) > > - mac(zuc) > > - sslmac(sha1) > > - shake128 > > - shake256 > > - cshake128 > > - cshake256 > > - kcmac128 > > - kcmac256 > > - kcmacxof128 > > - kcmacxof256 > > - sslmac(md5) > > Algorithms that don't have a generic implementation shouldn't be included. > > - Eric
On Wed, Mar 06, 2024 at 09:16:31AM +0530, Pavitrakumar Managutte wrote: > > Algorithms that don't have a generic implementation shouldn't be included. > > > Hi Eric, > Yes we have tested this with CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y > > Also is it fine if those additional algos stay inside that disabled > CONFIG. SPAcc hardware > does support those algos. > > - PK > If you provide an option to build them into the driver, that still counts as them being there. I think they should just be left out entirely for now. - Eric
Sure Eric, I will remove them from the driver, clean up all the build errors and submit v1 of this patchset. - PK On Wed, Mar 6, 2024 at 9:30 AM Eric Biggers <ebiggers@kernel.org> wrote: > > On Wed, Mar 06, 2024 at 09:16:31AM +0530, Pavitrakumar Managutte wrote: > > > Algorithms that don't have a generic implementation shouldn't be included. > > > > > Hi Eric, > > Yes we have tested this with CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y > > > > Also is it fine if those additional algos stay inside that disabled > > CONFIG. SPAcc hardware > > does support those algos. > > > > - PK > > > > If you provide an option to build them into the driver, that still counts as > them being there. I think they should just be left out entirely for now. > > - Eric