lib/crypto: x86/sha256: Move static_call above kernel-mode FPU section

As I did for sha512_blocks(), reorganize x86's sha256_blocks() to be
just a static_call.  To achieve that, for each assembly function add a C
function that handles the kernel-mode FPU section and fallback.  While
this increases total code size slightly, the amount of code actually
executed on a given system does not increase, and it is slightly more
efficient since it eliminates the extra static_key.  It also makes the
assembly functions be called with standard direct calls instead of
static calls, eliminating the need for ANNOTATE_NOENDBR.

Acked-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20250704023958.73274-2-ebiggers@kernel.org
Signed-off-by: Eric Biggers <ebiggers@kernel.org>

lib/crypto/x86/sha256-avx-asm.S

@@ -48,7 +48,6 @@
 ########################################################################
 
 #include <linux/linkage.h>
-#include <linux/objtool.h>
 
 ## assume buffers not aligned
 #define    VMOVDQ vmovdqu
@@ -346,8 +345,6 @@ a = TMP_
 ########################################################################
 .text
 SYM_FUNC_START(sha256_transform_avx)
-	ANNOTATE_NOENDBR	# since this is called only via static_call
-
 	pushq   %rbx
 	pushq   %r12
 	pushq   %r13

lib/crypto/x86/sha256-avx2-asm.S

@@ -49,7 +49,6 @@
 ########################################################################
 
 #include <linux/linkage.h>
-#include <linux/objtool.h>
 
 ## assume buffers not aligned
 #define	VMOVDQ vmovdqu
@@ -523,8 +522,6 @@ STACK_SIZE	= _CTX      + _CTX_SIZE
 ########################################################################
 .text
 SYM_FUNC_START(sha256_transform_rorx)
-	ANNOTATE_NOENDBR	# since this is called only via static_call
-
 	pushq	%rbx
 	pushq	%r12
 	pushq	%r13

lib/crypto/x86/sha256-ni-asm.S

@@ -54,7 +54,6 @@
  */
 
 #include <linux/linkage.h>
-#include <linux/objtool.h>
 
 #define STATE_PTR	%rdi	/* 1st arg */
 #define DATA_PTR	%rsi	/* 2nd arg */
@@ -111,7 +110,6 @@
  */
 .text
 SYM_FUNC_START(sha256_ni_transform)
-	ANNOTATE_NOENDBR	# since this is called only via static_call
 
 	shl		$6, NUM_BLKS		/*  convert to bytes */
 	jz		.Ldone_hash

lib/crypto/x86/sha256-ssse3-asm.S

@@ -47,7 +47,6 @@
 ########################################################################
 
 #include <linux/linkage.h>
-#include <linux/objtool.h>
 
 ## assume buffers not aligned
 #define    MOVDQ movdqu
@@ -353,8 +352,6 @@ a = TMP_
 ########################################################################
 .text
 SYM_FUNC_START(sha256_transform_ssse3)
-	ANNOTATE_NOENDBR	# since this is called only via static_call
-
 	pushq   %rbx
 	pushq   %r12
 	pushq   %r13

lib/crypto/x86/sha256.h

@@ -8,48 +8,50 @@
 #include <crypto/internal/simd.h>
 #include <linux/static_call.h>
 
-asmlinkage void sha256_transform_ssse3(struct sha256_block_state *state,
+DEFINE_STATIC_CALL(sha256_blocks_x86, sha256_blocks_generic);
-				       const u8 *data, size_t nblocks);
-asmlinkage void sha256_transform_avx(struct sha256_block_state *state,
-				     const u8 *data, size_t nblocks);
-asmlinkage void sha256_transform_rorx(struct sha256_block_state *state,
-				      const u8 *data, size_t nblocks);
-asmlinkage void sha256_ni_transform(struct sha256_block_state *state,
-				    const u8 *data, size_t nblocks);
 
-static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_sha256_x86);
+#define DEFINE_X86_SHA256_FN(c_fn, asm_fn)                                 \
+	asmlinkage void asm_fn(struct sha256_block_state *state,           \
+			       const u8 *data, size_t nblocks);            \
+	static void c_fn(struct sha256_block_state *state, const u8 *data, \
+			 size_t nblocks)                                   \
+	{                                                                  \
+		if (likely(crypto_simd_usable())) {                        \
+			kernel_fpu_begin();                                \
+			asm_fn(state, data, nblocks);                      \
+			kernel_fpu_end();                                  \
+		} else {                                                   \
+			sha256_blocks_generic(state, data, nblocks);       \
+		}                                                          \
+	}
 
-DEFINE_STATIC_CALL(sha256_blocks_x86, sha256_transform_ssse3);
+DEFINE_X86_SHA256_FN(sha256_blocks_ssse3, sha256_transform_ssse3);
+DEFINE_X86_SHA256_FN(sha256_blocks_avx, sha256_transform_avx);
+DEFINE_X86_SHA256_FN(sha256_blocks_avx2, sha256_transform_rorx);
+DEFINE_X86_SHA256_FN(sha256_blocks_ni, sha256_ni_transform);
 
 static void sha256_blocks(struct sha256_block_state *state,
 			  const u8 *data, size_t nblocks)
 {
-	if (static_branch_likely(&have_sha256_x86) && crypto_simd_usable()) {
+	static_call(sha256_blocks_x86)(state, data, nblocks);
-		kernel_fpu_begin();
-		static_call(sha256_blocks_x86)(state, data, nblocks);
-		kernel_fpu_end();
-	} else {
-		sha256_blocks_generic(state, data, nblocks);
-	}
 }
 
 #define sha256_mod_init_arch sha256_mod_init_arch
 static inline void sha256_mod_init_arch(void)
 {
 	if (boot_cpu_has(X86_FEATURE_SHA_NI)) {
-		static_call_update(sha256_blocks_x86, sha256_ni_transform);
+		static_call_update(sha256_blocks_x86, sha256_blocks_ni);
 	} else if (cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM,
 				     NULL) &&
 		   boot_cpu_has(X86_FEATURE_AVX)) {
 		if (boot_cpu_has(X86_FEATURE_AVX2) &&
 		    boot_cpu_has(X86_FEATURE_BMI2))
 			static_call_update(sha256_blocks_x86,
-					   sha256_transform_rorx);
+					   sha256_blocks_avx2);
 		else
 			static_call_update(sha256_blocks_x86,
-					   sha256_transform_avx);
+					   sha256_blocks_avx);
-	} else if (!boot_cpu_has(X86_FEATURE_SSSE3)) {
+	} else if (boot_cpu_has(X86_FEATURE_SSSE3)) {
-		return;
+		static_call_update(sha256_blocks_x86, sha256_blocks_ssse3);
 	}
-	static_branch_enable(&have_sha256_x86);
 }