Message ID | 1622202126-19237-3-git-send-email-yekai13@huawei.com (mailing list archive) |
---|---|
State | Accepted |
Delegated to: | Herbert Xu |
Headers | show |
Series | crypto: hisilicon - supports new skciphers for new hardware | expand |
Hi Kai, Thank you for the patch! Yet something to improve: [auto build test ERROR on cryptodev/master] [also build test ERROR on crypto/master linux/master linus/master v5.13-rc3 next-20210528] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in https://git-scm.com/docs/git-format-patch] url: https://github.com/0day-ci/linux/commits/Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 base: https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git master config: ia64-allmodconfig (attached as .config) compiler: ia64-linux-gcc (GCC) 9.3.0 reproduce (this is a W=1 build): wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross # https://github.com/0day-ci/linux/commit/60bae5ed49c53ea90c82125a8295fb72833a3b68 git remote add linux-review https://github.com/0day-ci/linux git fetch --no-tags linux-review Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 git checkout 60bae5ed49c53ea90c82125a8295fb72833a3b68 # save the attached .config to linux build tree COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross ARCH=ia64 If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot <lkp@intel.com> All errors (new ones prefixed by >>): drivers/crypto/hisilicon/sec2/sec_crypto.c: In function 'sec_aead_crypto': >> drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: error: 'sk_req' undeclared (first use in this function); did you mean 'a_req'? 1751 | return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); | ^~~~~~ | a_req drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: note: each undeclared identifier is reported only once for each function it appears in vim +1751 drivers/crypto/hisilicon/sec2/sec_crypto.c 1733 1734 static int sec_aead_crypto(struct aead_request *a_req, bool encrypt) 1735 { 1736 struct crypto_aead *tfm = crypto_aead_reqtfm(a_req); 1737 struct sec_req *req = aead_request_ctx(a_req); 1738 struct sec_ctx *ctx = crypto_aead_ctx(tfm); 1739 int ret; 1740 1741 req->flag = a_req->base.flags; 1742 req->aead_req.aead_req = a_req; 1743 req->c_req.encrypt = encrypt; 1744 req->ctx = ctx; 1745 1746 ret = sec_aead_param_check(ctx, req); 1747 if (unlikely(ret)) 1748 return -EINVAL; 1749 1750 if (unlikely(ctx->c_ctx.fallback)) > 1751 return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); 1752 1753 return ctx->req_op->process(ctx, req); 1754 } 1755 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
On 2021/5/28 22:13, kernel test robot wrote: > Hi Kai, > > Thank you for the patch! Yet something to improve: > > [auto build test ERROR on cryptodev/master] > [also build test ERROR on crypto/master linux/master linus/master v5.13-rc3 next-20210528] > [If your patch is applied to the wrong git tree, kindly drop us a note. > And when submitting patch, we suggest to use '--base' as documented in > https://git-scm.com/docs/git-format-patch] > > url: https://github.com/0day-ci/linux/commits/Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 > base: https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git master > config: ia64-allmodconfig (attached as .config) > compiler: ia64-linux-gcc (GCC) 9.3.0 > reproduce (this is a W=1 build): > wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross > chmod +x ~/bin/make.cross > # https://github.com/0day-ci/linux/commit/60bae5ed49c53ea90c82125a8295fb72833a3b68 > git remote add linux-review https://github.com/0day-ci/linux > git fetch --no-tags linux-review Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 > git checkout 60bae5ed49c53ea90c82125a8295fb72833a3b68 > # save the attached .config to linux build tree > COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross ARCH=ia64 > > If you fix the issue, kindly add following tag as appropriate > Reported-by: kernel test robot <lkp@intel.com> > > All errors (new ones prefixed by >>): > > drivers/crypto/hisilicon/sec2/sec_crypto.c: In function 'sec_aead_crypto': >>> drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: error: 'sk_req' undeclared (first use in this function); did you mean 'a_req'? > 1751 | return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); > | ^~~~~~ > | a_req > drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: note: each undeclared identifier is reported only once for each function it appears in > > > vim +1751 drivers/crypto/hisilicon/sec2/sec_crypto.c > > 1733 > 1734 static int sec_aead_crypto(struct aead_request *a_req, bool encrypt) > 1735 { > 1736 struct crypto_aead *tfm = crypto_aead_reqtfm(a_req); > 1737 struct sec_req *req = aead_request_ctx(a_req); > 1738 struct sec_ctx *ctx = crypto_aead_ctx(tfm); > 1739 int ret; > 1740 > 1741 req->flag = a_req->base.flags; > 1742 req->aead_req.aead_req = a_req; > 1743 req->c_req.encrypt = encrypt; > 1744 req->ctx = ctx; > 1745 > 1746 ret = sec_aead_param_check(ctx, req); > 1747 if (unlikely(ret)) > 1748 return -EINVAL; > 1749 > 1750 if (unlikely(ctx->c_ctx.fallback)) >> 1751 return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); > 1752 > 1753 return ctx->req_op->process(ctx, req); > 1754 } > 1755 > > --- > 0-DAY CI Kernel Test Service, Intel Corporation > https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org > You shouldn't git am this patchset directly, because this patchset depends on previous patchset. the series is "crypto: hisilicon - add new type of sqe for Kunpeng930", the patchwork is https://patchwork.kernel.org/project/linux-crypto/list/?series=490143, thank you Kai
On 5/29/21 5:22 PM, yekai(A) wrote: > > > On 2021/5/28 22:13, kernel test robot wrote: >> Hi Kai, >> >> Thank you for the patch! Yet something to improve: >> >> [auto build test ERROR on cryptodev/master] >> [also build test ERROR on crypto/master linux/master linus/master >> v5.13-rc3 next-20210528] >> [If your patch is applied to the wrong git tree, kindly drop us a note. >> And when submitting patch, we suggest to use '--base' as documented in >> https://git-scm.com/docs/git-format-patch] >> >> url: >> https://github.com/0day-ci/linux/commits/Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 >> base: >> https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git >> master >> config: ia64-allmodconfig (attached as .config) >> compiler: ia64-linux-gcc (GCC) 9.3.0 >> reproduce (this is a W=1 build): >> wget >> https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross >> -O ~/bin/make.cross >> chmod +x ~/bin/make.cross >> # >> https://github.com/0day-ci/linux/commit/60bae5ed49c53ea90c82125a8295fb72833a3b68 >> git remote add linux-review https://github.com/0day-ci/linux >> git fetch --no-tags linux-review >> Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 >> git checkout 60bae5ed49c53ea90c82125a8295fb72833a3b68 >> # save the attached .config to linux build tree >> COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 >> make.cross ARCH=ia64 >> >> If you fix the issue, kindly add following tag as appropriate >> Reported-by: kernel test robot <lkp@intel.com> >> >> All errors (new ones prefixed by >>): >> >> drivers/crypto/hisilicon/sec2/sec_crypto.c: In function >> 'sec_aead_crypto': >>>> drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: error: 'sk_req' >>>> undeclared (first use in this function); did you mean 'a_req'? >> 1751 | return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); >> | ^~~~~~ >> | a_req >> drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: note: each >> undeclared identifier is reported only once for each function it >> appears in >> >> >> vim +1751 drivers/crypto/hisilicon/sec2/sec_crypto.c >> >> 1733 >> 1734 static int sec_aead_crypto(struct aead_request *a_req, bool >> encrypt) >> 1735 { >> 1736 struct crypto_aead *tfm = crypto_aead_reqtfm(a_req); >> 1737 struct sec_req *req = aead_request_ctx(a_req); >> 1738 struct sec_ctx *ctx = crypto_aead_ctx(tfm); >> 1739 int ret; >> 1740 >> 1741 req->flag = a_req->base.flags; >> 1742 req->aead_req.aead_req = a_req; >> 1743 req->c_req.encrypt = encrypt; >> 1744 req->ctx = ctx; >> 1745 >> 1746 ret = sec_aead_param_check(ctx, req); >> 1747 if (unlikely(ret)) >> 1748 return -EINVAL; >> 1749 >> 1750 if (unlikely(ctx->c_ctx.fallback)) >>> 1751 return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); >> 1752 >> 1753 return ctx->req_op->process(ctx, req); >> 1754 } >> 1755 >> >> --- >> 0-DAY CI Kernel Test Service, Intel Corporation >> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org >> > > > You shouldn't git am this patchset directly, because this patchset > depends on previous patchset. > the series is "crypto: hisilicon - add new type of sqe for Kunpeng930", > the patchwork is > https://patchwork.kernel.org/project/linux-crypto/list/?series=490143, > thank you > Kai Hi Kai, Thanks for the clarification, it could be very helpful for us if '--base' for git format-patch used or there is a patchwork link in cover letter file. Best Regards, Rong Chen
On 2021/5/31 8:51, Rong Chen wrote: > > > On 5/29/21 5:22 PM, yekai(A) wrote: >> >> >> On 2021/5/28 22:13, kernel test robot wrote: >>> Hi Kai, >>> >>> Thank you for the patch! Yet something to improve: >>> >>> [auto build test ERROR on cryptodev/master] >>> [also build test ERROR on crypto/master linux/master linus/master >>> v5.13-rc3 next-20210528] >>> [If your patch is applied to the wrong git tree, kindly drop us a note. >>> And when submitting patch, we suggest to use '--base' as documented in >>> https://git-scm.com/docs/git-format-patch] >>> >>> url: >>> https://github.com/0day-ci/linux/commits/Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 >>> >>> base: >>> https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git >>> master >>> config: ia64-allmodconfig (attached as .config) >>> compiler: ia64-linux-gcc (GCC) 9.3.0 >>> reproduce (this is a W=1 build): >>> wget >>> https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross >>> -O ~/bin/make.cross >>> chmod +x ~/bin/make.cross >>> # >>> https://github.com/0day-ci/linux/commit/60bae5ed49c53ea90c82125a8295fb72833a3b68 >>> >>> git remote add linux-review https://github.com/0day-ci/linux >>> git fetch --no-tags linux-review >>> Kai-Ye/crypto-hisilicon-supports-new-skciphers-for-new-hardware/20210528-194644 >>> >>> git checkout 60bae5ed49c53ea90c82125a8295fb72833a3b68 >>> # save the attached .config to linux build tree >>> COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 >>> make.cross ARCH=ia64 >>> >>> If you fix the issue, kindly add following tag as appropriate >>> Reported-by: kernel test robot <lkp@intel.com> >>> >>> All errors (new ones prefixed by >>): >>> >>> drivers/crypto/hisilicon/sec2/sec_crypto.c: In function >>> 'sec_aead_crypto': >>>>> drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: error: 'sk_req' >>>>> undeclared (first use in this function); did you mean 'a_req'? >>> 1751 | return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); >>> | ^~~~~~ >>> | a_req >>> drivers/crypto/hisilicon/sec2/sec_crypto.c:1751:40: note: each >>> undeclared identifier is reported only once for each function it >>> appears in >>> >>> >>> vim +1751 drivers/crypto/hisilicon/sec2/sec_crypto.c >>> >>> 1733 >>> 1734 static int sec_aead_crypto(struct aead_request *a_req, bool >>> encrypt) >>> 1735 { >>> 1736 struct crypto_aead *tfm = crypto_aead_reqtfm(a_req); >>> 1737 struct sec_req *req = aead_request_ctx(a_req); >>> 1738 struct sec_ctx *ctx = crypto_aead_ctx(tfm); >>> 1739 int ret; >>> 1740 >>> 1741 req->flag = a_req->base.flags; >>> 1742 req->aead_req.aead_req = a_req; >>> 1743 req->c_req.encrypt = encrypt; >>> 1744 req->ctx = ctx; >>> 1745 >>> 1746 ret = sec_aead_param_check(ctx, req); >>> 1747 if (unlikely(ret)) >>> 1748 return -EINVAL; >>> 1749 >>> 1750 if (unlikely(ctx->c_ctx.fallback)) >>>> 1751 return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); >>> 1752 >>> 1753 return ctx->req_op->process(ctx, req); >>> 1754 } >>> 1755 >>> >>> --- >>> 0-DAY CI Kernel Test Service, Intel Corporation >>> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org >>> >> >> >> You shouldn't git am this patchset directly, because this patchset >> depends on previous patchset. >> the series is "crypto: hisilicon - add new type of sqe for Kunpeng930", >> the patchwork is >> https://patchwork.kernel.org/project/linux-crypto/list/?series=490143, >> thank you >> Kai > > Hi Kai, > > Thanks for the clarification, it could be very helpful for us if > '--base' for git format-patch used or > there is a patchwork link in cover letter file. > > Best Regards, > Rong Chen > . > thanks for your suggestion. Best Regards Kai
diff --git a/drivers/crypto/hisilicon/sec2/sec.h b/drivers/crypto/hisilicon/sec2/sec.h index 14ba66d..935d8d9 100644 --- a/drivers/crypto/hisilicon/sec2/sec.h +++ b/drivers/crypto/hisilicon/sec2/sec.h @@ -97,6 +97,10 @@ struct sec_cipher_ctx { u8 c_mode; u8 c_alg; u8 c_key_len; + + /* add software support */ + bool fallback; + struct crypto_sync_skcipher *fbtfm; }; /* SEC queue context which defines queue's relatives */ diff --git a/drivers/crypto/hisilicon/sec2/sec_crypto.c b/drivers/crypto/hisilicon/sec2/sec_crypto.c index b399315..25da33a 100644 --- a/drivers/crypto/hisilicon/sec2/sec_crypto.c +++ b/drivers/crypto/hisilicon/sec2/sec_crypto.c @@ -21,6 +21,7 @@ #define SEC_PRIORITY 4001 #define SEC_XTS_MIN_KEY_SIZE (2 * AES_MIN_KEY_SIZE) +#define SEC_XTS_MID_KEY_SIZE (3 * AES_MIN_KEY_SIZE) #define SEC_XTS_MAX_KEY_SIZE (2 * AES_MAX_KEY_SIZE) #define SEC_DES3_2KEY_SIZE (2 * DES_KEY_SIZE) #define SEC_DES3_3KEY_SIZE (3 * DES_KEY_SIZE) @@ -81,6 +82,7 @@ #define MAX_INPUT_DATA_LEN 0xFFFE00 #define BITS_MASK 0xFF #define BYTE_BITS 0x8 +#define SEC_XTS_NAME_SZ 0x3 /* Get an en/de-cipher queue cyclically to balance load over queues of TFM */ static inline int sec_alloc_queue_id(struct sec_ctx *ctx, struct sec_req *req) @@ -598,6 +600,26 @@ static void sec_auth_uninit(struct sec_ctx *ctx) a_ctx->a_key, a_ctx->a_key_dma); } +static int sec_skcipher_fbtfm_init(struct crypto_skcipher *tfm) +{ + const char *alg = crypto_tfm_alg_name(&tfm->base); + struct sec_ctx *ctx = crypto_skcipher_ctx(tfm); + struct sec_cipher_ctx *c_ctx = &ctx->c_ctx; + + c_ctx->fallback = false; + if (likely(strncmp(alg, "xts", SEC_XTS_NAME_SZ))) + return 0; + + c_ctx->fbtfm = crypto_alloc_sync_skcipher(alg, 0, + CRYPTO_ALG_NEED_FALLBACK); + if (IS_ERR(c_ctx->fbtfm)) { + pr_err("failed to alloc fallback tfm!\n"); + return PTR_ERR(c_ctx->fbtfm); + } + + return 0; +} + static int sec_skcipher_init(struct crypto_skcipher *tfm) { struct sec_ctx *ctx = crypto_skcipher_ctx(tfm); @@ -619,8 +641,14 @@ static int sec_skcipher_init(struct crypto_skcipher *tfm) if (ret) goto err_cipher_init; + ret = sec_skcipher_fbtfm_init(tfm); + if (ret) + goto err_fbtfm_init; + return 0; +err_fbtfm_init: + sec_cipher_uninit(ctx); err_cipher_init: sec_ctx_base_uninit(ctx); return ret; @@ -630,6 +658,9 @@ static void sec_skcipher_uninit(struct crypto_skcipher *tfm) { struct sec_ctx *ctx = crypto_skcipher_ctx(tfm); + if (ctx->c_ctx.fbtfm) + crypto_free_sync_skcipher(ctx->c_ctx.fbtfm); + sec_cipher_uninit(ctx); sec_ctx_base_uninit(ctx); } @@ -669,6 +700,9 @@ static int sec_skcipher_aes_sm4_setkey(struct sec_cipher_ctx *c_ctx, case SEC_XTS_MIN_KEY_SIZE: c_ctx->c_key_len = SEC_CKEY_128BIT; break; + case SEC_XTS_MID_KEY_SIZE: + c_ctx->fallback = true; + break; case SEC_XTS_MAX_KEY_SIZE: c_ctx->c_key_len = SEC_CKEY_256BIT; break; @@ -740,7 +774,13 @@ static int sec_skcipher_setkey(struct crypto_skcipher *tfm, const u8 *key, } memcpy(c_ctx->c_key, key, keylen); - + if (c_ctx->fallback) { + ret = crypto_sync_skcipher_setkey(c_ctx->fbtfm, key, keylen); + if (ret) { + dev_err(dev, "failed to set fallback skcipher key!\n"); + return ret; + } + } return 0; } @@ -1709,6 +1749,37 @@ static int sec_skcipher_param_check(struct sec_ctx *ctx, struct sec_req *sreq) return -EINVAL; } +static int sec_skcipher_soft_crypto(struct sec_ctx *ctx, + struct skcipher_request *sreq, bool encrypt) +{ + struct sec_cipher_ctx *c_ctx = &ctx->c_ctx; + struct device *dev = ctx->dev; + int ret; + + SYNC_SKCIPHER_REQUEST_ON_STACK(subreq, c_ctx->fbtfm); + + if (!c_ctx->fbtfm) { + dev_err(dev, "failed to check fallback tfm\n"); + return -EINVAL; + } + + skcipher_request_set_sync_tfm(subreq, c_ctx->fbtfm); + + /* software need sync mode to do crypto */ + skcipher_request_set_callback(subreq, sreq->base.flags, + NULL, NULL); + skcipher_request_set_crypt(subreq, sreq->src, sreq->dst, + sreq->cryptlen, sreq->iv); + if (encrypt) + ret = crypto_skcipher_encrypt(subreq); + else + ret = crypto_skcipher_decrypt(subreq); + + skcipher_request_zero(subreq); + + return ret; +} + static int sec_skcipher_crypto(struct skcipher_request *sk_req, bool encrypt) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(sk_req); @@ -1716,8 +1787,11 @@ static int sec_skcipher_crypto(struct skcipher_request *sk_req, bool encrypt) struct sec_ctx *ctx = crypto_skcipher_ctx(tfm); int ret; - if (!sk_req->cryptlen) + if (!sk_req->cryptlen) { + if (ctx->c_ctx.c_mode == SEC_CMODE_XTS) + return -EINVAL; return 0; + } req->flag = sk_req->base.flags; req->c_req.sk_req = sk_req; @@ -1728,6 +1802,9 @@ static int sec_skcipher_crypto(struct skcipher_request *sk_req, bool encrypt) if (unlikely(ret)) return -EINVAL; + if (unlikely(ctx->c_ctx.fallback)) + return sec_skcipher_soft_crypto(ctx, sk_req, encrypt); + return ctx->req_op->process(ctx, req); } @@ -1748,7 +1825,9 @@ static int sec_skcipher_decrypt(struct skcipher_request *sk_req) .cra_name = sec_cra_name,\ .cra_driver_name = "hisi_sec_"sec_cra_name,\ .cra_priority = SEC_PRIORITY,\ - .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY,\ + .cra_flags = CRYPTO_ALG_ASYNC |\ + CRYPTO_ALG_ALLOCATES_MEMORY |\ + CRYPTO_ALG_NEED_FALLBACK,\ .cra_blocksize = blk_size,\ .cra_ctxsize = sizeof(struct sec_ctx),\ .cra_module = THIS_MODULE,\