Merge pull request #8672 from SparkiDev/asm_fixes_1

Various fixes for Aarch64/ARM32/Thumb2 ASM

Author: JacobBarthelmeh

wolfcrypt/src/cpuid.c

@@ -187,6 +187,7 @@
                 cpuid_flags |= CPUID_SM4;
         #endif
 
+            (void)hwcaps;
             cpuid_check = 1;
         }
     }

wolfcrypt/src/port/arm/thumb2-aes-asm_c.c

@@ -103,7 +103,7 @@ void wc_chacha_setkey(word32* x, const byte* key, word32 keySz)
         "SUBS	%[keySz], %[keySz], #0x10\n\t"
         "ADD	r7, r7, %[keySz]\n\t"
         /* Start state with constants */
-        "ldm   r7, {r3, r4, r5, r6}\n\t"
+        "LDM	r7, {r3, r4, r5, r6}\n\t"
         "STM	%[x]!, {r3, r4, r5, r6}\n\t"
         /* Next is first 16 bytes of key. */
         "LDR	r3, [%[key]]\n\t"
@@ -176,7 +176,7 @@ void wc_chacha_crypt_bytes(ChaCha* ctx, byte* c, const byte* m, word32 len)
         "STRD	r4, r5, [sp, #16]\n\t"
         "STRD	r6, r7, [sp, #24]\n\t"
         /* Load x[0]..x[12] into registers. */
-        "ldm   lr, {r0, r1, r2, r3, r4, r5, r6, r7, r8, r9, r10, r11, r12}\n\t"
+        "LDM	lr, {r0, r1, r2, r3, r4, r5, r6, r7, r8, r9, r10, r11, r12}\n\t"
         /* 10x 2 full rounds to perform. */
         "MOV	lr, #0xa\n\t"
         "STR	lr, [sp, #48]\n\t"
@@ -315,35 +315,35 @@ void wc_chacha_crypt_bytes(ChaCha* ctx, byte* c, const byte* m, word32 len)
         "LDR	lr, [sp, #32]\n\t"
         "MOV	r12, sp\n\t"
         /* Add in original state */
-        "ldm   lr!, {r8, r9, r10, r11}\n\t"
+        "LDM	lr!, {r8, r9, r10, r11}\n\t"
         "ADD	%[ctx], %[ctx], r8\n\t"
         "ADD	%[c], %[c], r9\n\t"
         "ADD	%[m], %[m], r10\n\t"
         "ADD	%[len], %[len], r11\n\t"
-        "ldm   lr!, {r8, r9, r10, r11}\n\t"
+        "LDM	lr!, {r8, r9, r10, r11}\n\t"
         "ADD	r4, r4, r8\n\t"
         "ADD	r5, r5, r9\n\t"
         "ADD	r6, r6, r10\n\t"
         "ADD	r7, r7, r11\n\t"
-        "ldm   r12, {r8, r9}\n\t"
-        "ldm   lr!, {r10, r11}\n\t"
+        "LDM	r12, {r8, r9}\n\t"
+        "LDM	lr!, {r10, r11}\n\t"
         "ADD	r8, r8, r10\n\t"
         "ADD	r9, r9, r11\n\t"
         "STM	r12!, {r8, r9}\n\t"
-        "ldm   r12, {r8, r9}\n\t"
-        "ldm   lr!, {r10, r11}\n\t"
+        "LDM	r12, {r8, r9}\n\t"
+        "LDM	lr!, {r10, r11}\n\t"
         "ADD	r8, r8, r10\n\t"
         "ADD	r9, r9, r11\n\t"
         "STM	r12!, {r8, r9}\n\t"
-        "ldm   r12, {r8, r9}\n\t"
-        "ldm   lr!, {r10, r11}\n\t"
+        "LDM	r12, {r8, r9}\n\t"
+        "LDM	lr!, {r10, r11}\n\t"
         "ADD	r8, r8, r10\n\t"
         "ADD	r9, r9, r11\n\t"
         "ADD	r10, r10, #0x1\n\t"
         "STM	r12!, {r8, r9}\n\t"
         "STR	r10, [lr, #-8]\n\t"
-        "ldm   r12, {r8, r9}\n\t"
-        "ldm   lr, {r10, r11}\n\t"
+        "LDM	r12, {r8, r9}\n\t"
+        "LDM	lr, {r10, r11}\n\t"
         "ADD	r8, r8, r10\n\t"
         "ADD	r9, r9, r11\n\t"
         "STM	r12, {r8, r9}\n\t"
@@ -447,7 +447,7 @@ void wc_chacha_crypt_bytes(ChaCha* ctx, byte* c, const byte* m, word32 len)
         "LDR	lr, [sp, #32]\n\t"
         "ADD	r12, lr, #0x44\n\t"
         "STM	r12!, {%[ctx], %[c], %[m], %[len], r4, r5, r6, r7}\n\t"
-        "ldm   sp, {r0, r1, r2, r3, r4, r5, r6, r7}\n\t"
+        "LDM	sp, {r0, r1, r2, r3, r4, r5, r6, r7}\n\t"
         "STM	r12, {%[ctx], %[c], %[m], %[len], r4, r5, r6, r7}\n\t"
         "LDRD	%[m], %[len], [sp, #40]\n\t"
         "LDR	%[c], [sp, #36]\n\t"
@@ -469,7 +469,7 @@ void wc_chacha_crypt_bytes(ChaCha* ctx, byte* c, const byte* m, word32 len)
         "BLT.N	L_chacha_thumb2_crypt_word_loop_%=\n\t"
 #endif
         /* 16 bytes of state XORed into message. */
-        "ldm   lr!, {r4, r5, r6, r7}\n\t"
+        "LDM	lr!, {r4, r5, r6, r7}\n\t"
         "LDR	r8, [%[m]]\n\t"
         "LDR	r9, [%[m], #4]\n\t"
         "LDR	r10, [%[m], #8]\n\t"