crypto: powerpc/poly1305 - Add SIMD fallback

Add a SIMD fallback path for poly1305-p10 by converting the 2^64
based hash state into a 2^44 base.  In order to ensure that the
generic fallback is actually 2^44, add ARCH_SUPPORTS_INT128 to
powerpc and make poly1305-p10 depend on it.

Fixes: ba8f8624fde2 ("crypto: poly1305-p10 - Glue code for optmized Poly1305 implementation for ppc64le")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

3 files changed, 40 insertions, 4 deletions

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 6722625a406a0..651e0c32957ad 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -173,6 +173,7 @@ config PPC
 	select ARCH_STACKWALK
 	select ARCH_SUPPORTS_ATOMIC_RMW
 	select ARCH_SUPPORTS_DEBUG_PAGEALLOC	if PPC_BOOK3S || PPC_8xx
+	select ARCH_SUPPORTS_INT128		if PPC64 && CC_HAS_INT128
 	select ARCH_USE_BUILTIN_BSWAP
 	select ARCH_USE_CMPXCHG_LOCKREF		if PPC64
 	select ARCH_USE_MEMTEST
diff --git a/arch/powerpc/lib/crypto/Kconfig b/arch/powerpc/lib/crypto/Kconfig
index ffa541ad6d5da..6761fdb6193c4 100644
--- a/arch/powerpc/lib/crypto/Kconfig
+++ b/arch/powerpc/lib/crypto/Kconfig
@@ -9,7 +9,7 @@ config CRYPTO_CHACHA20_P10
 
 config CRYPTO_POLY1305_P10
 	tristate
-	depends on PPC64 && CPU_LITTLE_ENDIAN && VSX
+	depends on PPC64 && CPU_LITTLE_ENDIAN && VSX && ARCH_SUPPORTS_INT128
 	default CRYPTO_LIB_POLY1305
 	select CRYPTO_ARCH_HAVE_LIB_POLY1305
 	select CRYPTO_LIB_POLY1305_GENERIC
diff --git a/arch/powerpc/lib/crypto/poly1305-p10-glue.c b/arch/powerpc/lib/crypto/poly1305-p10-glue.c
index 3f1664a724b65..280c10c48c533 100644
--- a/arch/powerpc/lib/crypto/poly1305-p10-glue.c
+++ b/arch/powerpc/lib/crypto/poly1305-p10-glue.c
@@ -6,6 +6,7 @@
  */
 #include <asm/switch_to.h>
 #include <crypto/internal/poly1305.h>
+#include <crypto/internal/simd.h>
 #include <linux/cpufeature.h>
 #include <linux/jump_label.h>
 #include <linux/kernel.h>
@@ -18,6 +19,11 @@ asmlinkage void poly1305_emit_64(const struct poly1305_state *state, const u32 n
 
 static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_p10);
 
+static inline bool is_state_base64(struct poly1305_block_state *state)
+{
+	return state->core_r.precomputed_s.r64[2];
+}
+
 static void vsx_begin(void)
 {
 	preempt_disable();
@@ -30,12 +36,35 @@ static void vsx_end(void)
 	preempt_enable();
 }
 
+static void convert_to_base2_44(struct poly1305_block_state *state)
+{
+	u8 raw_key[POLY1305_BLOCK_SIZE];
+	u64 h0, h1, h2;
+
+	if (!is_state_base64(state))
+		return;
+
+	state->core_r.precomputed_s.r64[2] = 0;
+	put_unaligned_le64(state->core_r.key.r64[0], raw_key + 0);
+	put_unaligned_le64(state->core_r.key.r64[1], raw_key + 8);
+	poly1305_core_setkey(&state->core_r, raw_key);
+
+	h0 = state->h.h64[0];
+	h1 = state->h.h64[1];
+	h2 = state->h.h64[2];
+	state->h.h64[0] = h0 & 0xfffffffffffULL;
+	state->h.h64[1] = h0 >> 44 | (h1 & 0xffffffULL) << 20;
+	state->h.h64[2] = h1 >> 24 | h2 << 40;
+}
+
 void poly1305_block_init_arch(struct poly1305_block_state *dctx,
 			      const u8 raw_key[POLY1305_BLOCK_SIZE])
 {
-	if (!static_key_enabled(&have_p10))
+	dctx->core_r.precomputed_s.r64[2] = 0;
+	if (!static_key_enabled(&have_p10) || !crypto_simd_usable())
 		return poly1305_block_init_generic(dctx, raw_key);
 
+	dctx->core_r.precomputed_s.r64[2] = 1;
 	dctx->h = (struct poly1305_state){};
 	dctx->core_r.key.r64[0] = get_unaligned_le64(raw_key + 0);
 	dctx->core_r.key.r64[1] = get_unaligned_le64(raw_key + 8);
@@ -45,8 +74,11 @@ EXPORT_SYMBOL_GPL(poly1305_block_init_arch);
 void poly1305_blocks_arch(struct poly1305_block_state *state, const u8 *src,
 			  unsigned int len, u32 padbit)
 {
-	if (!static_key_enabled(&have_p10))
+	if (!static_key_enabled(&have_p10) || !is_state_base64(state) ||
+	    !crypto_simd_usable()) {
+		convert_to_base2_44(state);
 		return poly1305_blocks_generic(state, src, len, padbit);
+	}
 	vsx_begin();
 	if (len >= POLY1305_BLOCK_SIZE * 4) {
 		poly1305_p10le_4blocks(state, src, len);
@@ -66,7 +98,10 @@ void poly1305_emit_arch(const struct poly1305_state *state,
 			u8 digest[POLY1305_DIGEST_SIZE],
 			const u32 nonce[4])
 {
-	if (!static_key_enabled(&have_p10))
+	struct poly1305_block_state *dctx =
+		container_of(state, struct poly1305_block_state, h);
+
+	if (!static_key_enabled(&have_p10) || !is_state_base64(dctx))
 		return poly1305_emit_generic(state, digest, nonce);
 	poly1305_emit_64(state, nonce, digest);
 }