From b4e1f8548ada116704f0f4ac4a5a5367d1651751 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Wed, 8 Jul 2020 12:11:18 +0300 Subject: [PATCH] UPSTREAM: crypto: x86/chacha-sse3 - use unaligned loads for state array Due to the fact that the x86 port does not support allocating objects on the stack with an alignment that exceeds 8 bytes, we have a rather ugly hack in the x86 code for ChaCha to ensure that the state array is aligned to 16 bytes, allowing the SSE3 implementation of the algorithm to use aligned loads. Given that the performance benefit of using of aligned loads appears to be limited (~0.25% for 1k blocks using tcrypt on a Corei7-8650U), and the fact that this hack has leaked into generic ChaCha code, let's just remove it. Cc: Martin Willi Cc: Herbert Xu Cc: Eric Biggers Signed-off-by: Ard Biesheuvel Reviewed-by: Martin Willi Reviewed-by: Eric Biggers Signed-off-by: Herbert Xu (cherry picked from commit e79a31715193686e92dadb4caedfbb1f5de3659c) Bug: 152722841 Signed-off-by: Jason A. Donenfeld Signed-off-by: Greg Kroah-Hartman Change-Id: Ica3b8a25817a27053e537280ca08caca817dc879 --- arch/x86/crypto/chacha-ssse3-x86_64.S | 16 ++++++++-------- arch/x86/crypto/chacha_glue.c | 17 ++--------------- include/crypto/chacha.h | 4 ---- 3 files changed, 10 insertions(+), 27 deletions(-) diff --git a/arch/x86/crypto/chacha-ssse3-x86_64.S b/arch/x86/crypto/chacha-ssse3-x86_64.S index 2d86c7d6dc88..a556a57f1a4a 100644 --- a/arch/x86/crypto/chacha-ssse3-x86_64.S +++ b/arch/x86/crypto/chacha-ssse3-x86_64.S @@ -120,10 +120,10 @@ ENTRY(chacha_block_xor_ssse3) FRAME_BEGIN # x0..3 = s0..3 - movdqa 0x00(%rdi),%xmm0 - movdqa 0x10(%rdi),%xmm1 - movdqa 0x20(%rdi),%xmm2 - movdqa 0x30(%rdi),%xmm3 + movdqu 0x00(%rdi),%xmm0 + movdqu 0x10(%rdi),%xmm1 + movdqu 0x20(%rdi),%xmm2 + movdqu 0x30(%rdi),%xmm3 movdqa %xmm0,%xmm8 movdqa %xmm1,%xmm9 movdqa %xmm2,%xmm10 @@ -205,10 +205,10 @@ ENTRY(hchacha_block_ssse3) # %edx: nrounds FRAME_BEGIN - movdqa 0x00(%rdi),%xmm0 - movdqa 0x10(%rdi),%xmm1 - movdqa 0x20(%rdi),%xmm2 - movdqa 0x30(%rdi),%xmm3 + movdqu 0x00(%rdi),%xmm0 + movdqu 0x10(%rdi),%xmm1 + movdqu 0x20(%rdi),%xmm2 + movdqu 0x30(%rdi),%xmm3 mov %edx,%r8d call chacha_permute diff --git a/arch/x86/crypto/chacha_glue.c b/arch/x86/crypto/chacha_glue.c index f3bfce21bc0d..4c4dc64398cb 100644 --- a/arch/x86/crypto/chacha_glue.c +++ b/arch/x86/crypto/chacha_glue.c @@ -14,8 +14,6 @@ #include #include -#define CHACHA_STATE_ALIGN 16 - asmlinkage void chacha_block_xor_ssse3(u32 *state, u8 *dst, const u8 *src, unsigned int len, int nrounds); asmlinkage void chacha_4block_xor_ssse3(u32 *state, u8 *dst, const u8 *src, @@ -125,8 +123,6 @@ static void chacha_dosimd(u32 *state, u8 *dst, const u8 *src, void hchacha_block_arch(const u32 *state, u32 *stream, int nrounds) { - state = PTR_ALIGN(state, CHACHA_STATE_ALIGN); - if (!static_branch_likely(&chacha_use_simd) || !crypto_simd_usable()) { hchacha_block_generic(state, stream, nrounds); } else { @@ -139,8 +135,6 @@ EXPORT_SYMBOL(hchacha_block_arch); void chacha_init_arch(u32 *state, const u32 *key, const u8 *iv) { - state = PTR_ALIGN(state, CHACHA_STATE_ALIGN); - chacha_init_generic(state, key, iv); } EXPORT_SYMBOL(chacha_init_arch); @@ -148,8 +142,6 @@ EXPORT_SYMBOL(chacha_init_arch); void chacha_crypt_arch(u32 *state, u8 *dst, const u8 *src, unsigned int bytes, int nrounds) { - state = PTR_ALIGN(state, CHACHA_STATE_ALIGN); - if (!static_branch_likely(&chacha_use_simd) || !crypto_simd_usable() || bytes <= CHACHA_BLOCK_SIZE) return chacha_crypt_generic(state, dst, src, bytes, nrounds); @@ -171,15 +163,12 @@ EXPORT_SYMBOL(chacha_crypt_arch); static int chacha_simd_stream_xor(struct skcipher_request *req, const struct chacha_ctx *ctx, const u8 *iv) { - u32 *state, state_buf[16 + 2] __aligned(8); + u32 state[CHACHA_STATE_WORDS] __aligned(8); struct skcipher_walk walk; int err; err = skcipher_walk_virt(&walk, req, false); - BUILD_BUG_ON(CHACHA_STATE_ALIGN != 16); - state = PTR_ALIGN(state_buf + 0, CHACHA_STATE_ALIGN); - chacha_init_generic(state, ctx->key, iv); while (walk.nbytes > 0) { @@ -218,12 +207,10 @@ static int xchacha_simd(struct skcipher_request *req) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct chacha_ctx *ctx = crypto_skcipher_ctx(tfm); - u32 *state, state_buf[16 + 2] __aligned(8); + u32 state[CHACHA_STATE_WORDS] __aligned(8); struct chacha_ctx subctx; u8 real_iv[16]; - BUILD_BUG_ON(CHACHA_STATE_ALIGN != 16); - state = PTR_ALIGN(state_buf + 0, CHACHA_STATE_ALIGN); chacha_init_generic(state, ctx->key, req->iv); if (req->cryptlen > CHACHA_BLOCK_SIZE && crypto_simd_usable()) { diff --git a/include/crypto/chacha.h b/include/crypto/chacha.h index d4bfc9cb770e..b73c9931f943 100644 --- a/include/crypto/chacha.h +++ b/include/crypto/chacha.h @@ -25,11 +25,7 @@ #define CHACHA_BLOCK_SIZE 64 #define CHACHAPOLY_IV_SIZE 12 -#ifdef CONFIG_X86_64 -#define CHACHA_STATE_WORDS ((CHACHA_BLOCK_SIZE + 12) / sizeof(u32)) -#else #define CHACHA_STATE_WORDS (CHACHA_BLOCK_SIZE / sizeof(u32)) -#endif /* 192-bit nonce, then 64-bit stream position */ #define XCHACHA_IV_SIZE 32