diff --git a/raid/x86.c b/raid/x86.c
index bf13842f..0ea251f1 100644
--- a/raid/x86.c
+++ b/raid/x86.c
@@ -36,20 +36,20 @@ void raid_gen1_sse2(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	for (i = 0; i < size; i += 64) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (v[l][i]));
-		asm volatile("movdqa %0,%%xmm1" : : "m" (v[l][i+16]));
-		asm volatile("movdqa %0,%%xmm2" : : "m" (v[l][i+32]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (v[l][i+48]));
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (v[l][i+16]));
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (v[l][i+32]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (v[l][i+48]));
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("pxor %0,%%xmm0" : : "m" (v[d][i]));
-			asm volatile("pxor %0,%%xmm1" : : "m" (v[d][i+16]));
-			asm volatile("pxor %0,%%xmm2" : : "m" (v[d][i+32]));
-			asm volatile("pxor %0,%%xmm3" : : "m" (v[d][i+48]));
+			asm volatile ("pxor %0,%%xmm0" : : "m" (v[d][i]));
+			asm volatile ("pxor %0,%%xmm1" : : "m" (v[d][i+16]));
+			asm volatile ("pxor %0,%%xmm2" : : "m" (v[d][i+32]));
+			asm volatile ("pxor %0,%%xmm3" : : "m" (v[d][i+48]));
 		}
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (p[i+16]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (p[i+32]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (p[i+48]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (p[i+16]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (p[i+32]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (p[i+48]));
 	}
 
 	raid_asm_clobber_xmm4();
@@ -79,14 +79,14 @@ void raid_gen1_avx2(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	for (i = 0; i < size; i += 64) {
-		asm volatile("vmovdqa %0,%%ymm0" : : "m" (v[l][i]));
-		asm volatile("vmovdqa %0,%%ymm1" : : "m" (v[l][i+32]));
+		asm volatile ("vmovdqa %0,%%ymm0" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm1" : : "m" (v[l][i+32]));
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("vpxor %0,%%ymm0,%%ymm0" : : "m" (v[d][i]));
-			asm volatile("vpxor %0,%%ymm1,%%ymm1" : : "m" (v[d][i+32]));
+			asm volatile ("vpxor %0,%%ymm0,%%ymm0" : : "m" (v[d][i]));
+			asm volatile ("vpxor %0,%%ymm1,%%ymm1" : : "m" (v[d][i+32]));
 		}
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (p[i+32]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (p[i+32]));
 	}
 
 	raid_asm_clobber_ymm4();
@@ -129,36 +129,36 @@ void raid_gen2_sse2(int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
 
 	for (i = 0; i < size; i += 32) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (v[l][i]));
-		asm volatile("movdqa %0,%%xmm1" : : "m" (v[l][i+16]));
-		asm volatile("movdqa %xmm0,%xmm2");
-		asm volatile("movdqa %xmm1,%xmm3");
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (v[l][i+16]));
+		asm volatile ("movdqa %xmm0,%xmm2");
+		asm volatile ("movdqa %xmm1,%xmm3");
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("pxor %xmm4,%xmm4");
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pcmpgtb %xmm2,%xmm4");
-			asm volatile("pcmpgtb %xmm3,%xmm5");
-			asm volatile("paddb %xmm2,%xmm2");
-			asm volatile("paddb %xmm3,%xmm3");
-			asm volatile("pand %xmm7,%xmm4");
-			asm volatile("pand %xmm7,%xmm5");
-			asm volatile("pxor %xmm4,%xmm2");
-			asm volatile("pxor %xmm5,%xmm3");
+			asm volatile ("pxor %xmm4,%xmm4");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pcmpgtb %xmm2,%xmm4");
+			asm volatile ("pcmpgtb %xmm3,%xmm5");
+			asm volatile ("paddb %xmm2,%xmm2");
+			asm volatile ("paddb %xmm3,%xmm3");
+			asm volatile ("pand %xmm7,%xmm4");
+			asm volatile ("pand %xmm7,%xmm5");
+			asm volatile ("pxor %xmm4,%xmm2");
+			asm volatile ("pxor %xmm5,%xmm3");
 
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
-			asm volatile("movdqa %0,%%xmm5" : : "m" (v[d][i+16]));
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm5,%xmm1");
-			asm volatile("pxor %xmm4,%xmm2");
-			asm volatile("pxor %xmm5,%xmm3");
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm5" : : "m" (v[d][i+16]));
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm5,%xmm1");
+			asm volatile ("pxor %xmm4,%xmm2");
+			asm volatile ("pxor %xmm5,%xmm3");
 		}
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (p[i+16]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (q[i+16]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (p[i+16]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (q[i+16]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -185,35 +185,35 @@ void raid_gen2_avx2(int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("vbroadcasti128 %0, %%ymm7" : : "m" (gfconst16.poly[0]));
-	asm volatile("vpxor %ymm6,%ymm6,%ymm6");
+	asm volatile ("vbroadcasti128 %0, %%ymm7" : : "m" (gfconst16.poly[0]));
+	asm volatile ("vpxor %ymm6,%ymm6,%ymm6");
 
 	for (i = 0; i < size; i += 64) {
-		asm volatile("vmovdqa %0,%%ymm0" : : "m" (v[l][i]));
-		asm volatile("vmovdqa %0,%%ymm1" : : "m" (v[l][i+32]));
-		asm volatile("vmovdqa %ymm0,%ymm2");
-		asm volatile("vmovdqa %ymm1,%ymm3");
+		asm volatile ("vmovdqa %0,%%ymm0" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm1" : : "m" (v[l][i+32]));
+		asm volatile ("vmovdqa %ymm0,%ymm2");
+		asm volatile ("vmovdqa %ymm1,%ymm3");
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("vpcmpgtb %ymm2,%ymm6,%ymm4");
-			asm volatile("vpcmpgtb %ymm3,%ymm6,%ymm5");
-			asm volatile("vpaddb %ymm2,%ymm2,%ymm2");
-			asm volatile("vpaddb %ymm3,%ymm3,%ymm3");
-			asm volatile("vpand %ymm7,%ymm4,%ymm4");
-			asm volatile("vpand %ymm7,%ymm5,%ymm5");
-			asm volatile("vpxor %ymm4,%ymm2,%ymm2");
-			asm volatile("vpxor %ymm5,%ymm3,%ymm3");
+			asm volatile ("vpcmpgtb %ymm2,%ymm6,%ymm4");
+			asm volatile ("vpcmpgtb %ymm3,%ymm6,%ymm5");
+			asm volatile ("vpaddb %ymm2,%ymm2,%ymm2");
+			asm volatile ("vpaddb %ymm3,%ymm3,%ymm3");
+			asm volatile ("vpand %ymm7,%ymm4,%ymm4");
+			asm volatile ("vpand %ymm7,%ymm5,%ymm5");
+			asm volatile ("vpxor %ymm4,%ymm2,%ymm2");
+			asm volatile ("vpxor %ymm5,%ymm3,%ymm3");
 
-			asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
-			asm volatile("vmovdqa %0,%%ymm5" : : "m" (v[d][i+32]));
-			asm volatile("vpxor %ymm4,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm5,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm4,%ymm2,%ymm2");
-			asm volatile("vpxor %ymm5,%ymm3,%ymm3");
+			asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
+			asm volatile ("vmovdqa %0,%%ymm5" : : "m" (v[d][i+32]));
+			asm volatile ("vpxor %ymm4,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm5,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm4,%ymm2,%ymm2");
+			asm volatile ("vpxor %ymm5,%ymm3,%ymm3");
 		}
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (p[i+32]));
-		asm volatile("vmovntdq %%ymm2,%0" : "=m" (q[i]));
-		asm volatile("vmovntdq %%ymm3,%0" : "=m" (q[i+32]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (p[i+32]));
+		asm volatile ("vmovntdq %%ymm2,%0" : "=m" (q[i]));
+		asm volatile ("vmovntdq %%ymm3,%0" : "=m" (q[i+32]));
 	}
 
 	raid_asm_clobber_ymm8();
@@ -242,60 +242,60 @@ void raid_gen2_sse2ext(int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm15" : : "m" (gfconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm15" : : "m" (gfconst16.poly[0]));
 
 	for (i = 0; i < size; i += 64) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (v[l][i]));
-		asm volatile("movdqa %0,%%xmm1" : : "m" (v[l][i+16]));
-		asm volatile("movdqa %0,%%xmm2" : : "m" (v[l][i+32]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (v[l][i+48]));
-		asm volatile("movdqa %xmm0,%xmm4");
-		asm volatile("movdqa %xmm1,%xmm5");
-		asm volatile("movdqa %xmm2,%xmm6");
-		asm volatile("movdqa %xmm3,%xmm7");
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (v[l][i+16]));
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (v[l][i+32]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (v[l][i+48]));
+		asm volatile ("movdqa %xmm0,%xmm4");
+		asm volatile ("movdqa %xmm1,%xmm5");
+		asm volatile ("movdqa %xmm2,%xmm6");
+		asm volatile ("movdqa %xmm3,%xmm7");
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("pxor %xmm8,%xmm8");
-			asm volatile("pxor %xmm9,%xmm9");
-			asm volatile("pxor %xmm10,%xmm10");
-			asm volatile("pxor %xmm11,%xmm11");
-			asm volatile("pcmpgtb %xmm4,%xmm8");
-			asm volatile("pcmpgtb %xmm5,%xmm9");
-			asm volatile("pcmpgtb %xmm6,%xmm10");
-			asm volatile("pcmpgtb %xmm7,%xmm11");
-			asm volatile("paddb %xmm4,%xmm4");
-			asm volatile("paddb %xmm5,%xmm5");
-			asm volatile("paddb %xmm6,%xmm6");
-			asm volatile("paddb %xmm7,%xmm7");
-			asm volatile("pand %xmm15,%xmm8");
-			asm volatile("pand %xmm15,%xmm9");
-			asm volatile("pand %xmm15,%xmm10");
-			asm volatile("pand %xmm15,%xmm11");
-			asm volatile("pxor %xmm8,%xmm4");
-			asm volatile("pxor %xmm9,%xmm5");
-			asm volatile("pxor %xmm10,%xmm6");
-			asm volatile("pxor %xmm11,%xmm7");
+			asm volatile ("pxor %xmm8,%xmm8");
+			asm volatile ("pxor %xmm9,%xmm9");
+			asm volatile ("pxor %xmm10,%xmm10");
+			asm volatile ("pxor %xmm11,%xmm11");
+			asm volatile ("pcmpgtb %xmm4,%xmm8");
+			asm volatile ("pcmpgtb %xmm5,%xmm9");
+			asm volatile ("pcmpgtb %xmm6,%xmm10");
+			asm volatile ("pcmpgtb %xmm7,%xmm11");
+			asm volatile ("paddb %xmm4,%xmm4");
+			asm volatile ("paddb %xmm5,%xmm5");
+			asm volatile ("paddb %xmm6,%xmm6");
+			asm volatile ("paddb %xmm7,%xmm7");
+			asm volatile ("pand %xmm15,%xmm8");
+			asm volatile ("pand %xmm15,%xmm9");
+			asm volatile ("pand %xmm15,%xmm10");
+			asm volatile ("pand %xmm15,%xmm11");
+			asm volatile ("pxor %xmm8,%xmm4");
+			asm volatile ("pxor %xmm9,%xmm5");
+			asm volatile ("pxor %xmm10,%xmm6");
+			asm volatile ("pxor %xmm11,%xmm7");
 
-			asm volatile("movdqa %0,%%xmm8" : : "m" (v[d][i]));
-			asm volatile("movdqa %0,%%xmm9" : : "m" (v[d][i+16]));
-			asm volatile("movdqa %0,%%xmm10" : : "m" (v[d][i+32]));
-			asm volatile("movdqa %0,%%xmm11" : : "m" (v[d][i+48]));
-			asm volatile("pxor %xmm8,%xmm0");
-			asm volatile("pxor %xmm9,%xmm1");
-			asm volatile("pxor %xmm10,%xmm2");
-			asm volatile("pxor %xmm11,%xmm3");
-			asm volatile("pxor %xmm8,%xmm4");
-			asm volatile("pxor %xmm9,%xmm5");
-			asm volatile("pxor %xmm10,%xmm6");
-			asm volatile("pxor %xmm11,%xmm7");
+			asm volatile ("movdqa %0,%%xmm8" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm9" : : "m" (v[d][i+16]));
+			asm volatile ("movdqa %0,%%xmm10" : : "m" (v[d][i+32]));
+			asm volatile ("movdqa %0,%%xmm11" : : "m" (v[d][i+48]));
+			asm volatile ("pxor %xmm8,%xmm0");
+			asm volatile ("pxor %xmm9,%xmm1");
+			asm volatile ("pxor %xmm10,%xmm2");
+			asm volatile ("pxor %xmm11,%xmm3");
+			asm volatile ("pxor %xmm8,%xmm4");
+			asm volatile ("pxor %xmm9,%xmm5");
+			asm volatile ("pxor %xmm10,%xmm6");
+			asm volatile ("pxor %xmm11,%xmm7");
 		}
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (p[i+16]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (p[i+32]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (p[i+48]));
-		asm volatile("movntdq %%xmm4,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm5,%0" : "=m" (q[i+16]));
-		asm volatile("movntdq %%xmm6,%0" : "=m" (q[i+32]));
-		asm volatile("movntdq %%xmm7,%0" : "=m" (q[i+48]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (p[i+16]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (p[i+32]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (p[i+48]));
+		asm volatile ("movntdq %%xmm4,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm5,%0" : "=m" (q[i+16]));
+		asm volatile ("movntdq %%xmm6,%0" : "=m" (q[i+32]));
+		asm volatile ("movntdq %%xmm7,%0" : "=m" (q[i+48]));
 	}
 
 	raid_asm_clobber_xmm16();
@@ -332,69 +332,69 @@ void raid_gen3_ssse3(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("movdqa %0,%%xmm3" : : "m" (gfconst16.poly[0]));
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm3" : : "m" (gfconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 16) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[l][i]));
 
-		asm volatile("movdqa %xmm4,%xmm0");
-		asm volatile("movdqa %xmm4,%xmm1");
+		asm volatile ("movdqa %xmm4,%xmm0");
+		asm volatile ("movdqa %xmm4,%xmm1");
 
-		asm volatile("movdqa %xmm4,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
+		asm volatile ("movdqa %xmm4,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm6");
-		asm volatile("pxor   %xmm6,%xmm2");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm6");
+		asm volatile ("pxor   %xmm6,%xmm2");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
 
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pcmpgtb %xmm1,%xmm5");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("pand %xmm3,%xmm5");
-			asm volatile("pxor %xmm5,%xmm1");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pcmpgtb %xmm1,%xmm5");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("pand %xmm3,%xmm5");
+			asm volatile ("pxor %xmm5,%xmm1");
 
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm1");
 
-			asm volatile("movdqa %xmm4,%xmm5");
-			asm volatile("psrlw  $4,%xmm5");
-			asm volatile("pand   %xmm7,%xmm4");
-			asm volatile("pand   %xmm7,%xmm5");
+			asm volatile ("movdqa %xmm4,%xmm5");
+			asm volatile ("psrlw  $4,%xmm5");
+			asm volatile ("pand   %xmm7,%xmm4");
+			asm volatile ("pand   %xmm7,%xmm5");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pxor   %xmm6,%xmm2");
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("pshufb %xmm5,%xmm6");
-			asm volatile("pxor   %xmm6,%xmm2");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pxor   %xmm6,%xmm2");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("pshufb %xmm5,%xmm6");
+			asm volatile ("pxor   %xmm6,%xmm2");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[0][i]));
 
-		asm volatile("pxor %xmm5,%xmm5");
-		asm volatile("pcmpgtb %xmm1,%xmm5");
-		asm volatile("paddb %xmm1,%xmm1");
-		asm volatile("pand %xmm3,%xmm5");
-		asm volatile("pxor %xmm5,%xmm1");
+		asm volatile ("pxor %xmm5,%xmm5");
+		asm volatile ("pcmpgtb %xmm1,%xmm5");
+		asm volatile ("paddb %xmm1,%xmm1");
+		asm volatile ("pand %xmm3,%xmm5");
+		asm volatile ("pxor %xmm5,%xmm1");
 
-		asm volatile("pxor %xmm4,%xmm0");
-		asm volatile("pxor %xmm4,%xmm1");
-		asm volatile("pxor %xmm4,%xmm2");
+		asm volatile ("pxor %xmm4,%xmm0");
+		asm volatile ("pxor %xmm4,%xmm1");
+		asm volatile ("pxor %xmm4,%xmm2");
 
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -433,111 +433,111 @@ void raid_gen3_ssse3ext(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("movdqa %0,%%xmm3" : : "m" (gfconst16.poly[0]));
-	asm volatile("movdqa %0,%%xmm11" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm3" : : "m" (gfconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm11" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 32) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[l][i]));
-		asm volatile("movdqa %0,%%xmm12" : : "m" (v[l][i+16]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm12" : : "m" (v[l][i+16]));
 
-		asm volatile("movdqa %xmm4,%xmm0");
-		asm volatile("movdqa %xmm4,%xmm1");
-		asm volatile("movdqa %xmm12,%xmm8");
-		asm volatile("movdqa %xmm12,%xmm9");
+		asm volatile ("movdqa %xmm4,%xmm0");
+		asm volatile ("movdqa %xmm4,%xmm1");
+		asm volatile ("movdqa %xmm12,%xmm8");
+		asm volatile ("movdqa %xmm12,%xmm9");
 
-		asm volatile("movdqa %xmm4,%xmm5");
-		asm volatile("movdqa %xmm12,%xmm13");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("psrlw  $4,%xmm13");
-		asm volatile("pand   %xmm11,%xmm4");
-		asm volatile("pand   %xmm11,%xmm12");
-		asm volatile("pand   %xmm11,%xmm5");
-		asm volatile("pand   %xmm11,%xmm13");
+		asm volatile ("movdqa %xmm4,%xmm5");
+		asm volatile ("movdqa %xmm12,%xmm13");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("psrlw  $4,%xmm13");
+		asm volatile ("pand   %xmm11,%xmm4");
+		asm volatile ("pand   %xmm11,%xmm12");
+		asm volatile ("pand   %xmm11,%xmm5");
+		asm volatile ("pand   %xmm11,%xmm13");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("movdqa %xmm2,%xmm10");
-		asm volatile("movdqa %xmm7,%xmm15");
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm12,%xmm10");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pshufb %xmm13,%xmm15");
-		asm volatile("pxor   %xmm7,%xmm2");
-		asm volatile("pxor   %xmm15,%xmm10");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("movdqa %xmm2,%xmm10");
+		asm volatile ("movdqa %xmm7,%xmm15");
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm12,%xmm10");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pshufb %xmm13,%xmm15");
+		asm volatile ("pxor   %xmm7,%xmm2");
+		asm volatile ("pxor   %xmm15,%xmm10");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
-			asm volatile("movdqa %0,%%xmm12" : : "m" (v[d][i+16]));
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (v[d][i+16]));
 
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pxor %xmm13,%xmm13");
-			asm volatile("pcmpgtb %xmm1,%xmm5");
-			asm volatile("pcmpgtb %xmm9,%xmm13");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("paddb %xmm9,%xmm9");
-			asm volatile("pand %xmm3,%xmm5");
-			asm volatile("pand %xmm3,%xmm13");
-			asm volatile("pxor %xmm5,%xmm1");
-			asm volatile("pxor %xmm13,%xmm9");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pxor %xmm13,%xmm13");
+			asm volatile ("pcmpgtb %xmm1,%xmm5");
+			asm volatile ("pcmpgtb %xmm9,%xmm13");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("paddb %xmm9,%xmm9");
+			asm volatile ("pand %xmm3,%xmm5");
+			asm volatile ("pand %xmm3,%xmm13");
+			asm volatile ("pxor %xmm5,%xmm1");
+			asm volatile ("pxor %xmm13,%xmm9");
 
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm1");
-			asm volatile("pxor %xmm12,%xmm8");
-			asm volatile("pxor %xmm12,%xmm9");
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm12,%xmm8");
+			asm volatile ("pxor %xmm12,%xmm9");
 
-			asm volatile("movdqa %xmm4,%xmm5");
-			asm volatile("movdqa %xmm12,%xmm13");
-			asm volatile("psrlw  $4,%xmm5");
-			asm volatile("psrlw  $4,%xmm13");
-			asm volatile("pand   %xmm11,%xmm4");
-			asm volatile("pand   %xmm11,%xmm12");
-			asm volatile("pand   %xmm11,%xmm5");
-			asm volatile("pand   %xmm11,%xmm13");
+			asm volatile ("movdqa %xmm4,%xmm5");
+			asm volatile ("movdqa %xmm12,%xmm13");
+			asm volatile ("psrlw  $4,%xmm5");
+			asm volatile ("psrlw  $4,%xmm13");
+			asm volatile ("pand   %xmm11,%xmm4");
+			asm volatile ("pand   %xmm11,%xmm12");
+			asm volatile ("pand   %xmm11,%xmm5");
+			asm volatile ("pand   %xmm11,%xmm13");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("movdqa %xmm6,%xmm14");
-			asm volatile("movdqa %xmm7,%xmm15");
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm12,%xmm14");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pshufb %xmm13,%xmm15");
-			asm volatile("pxor   %xmm6,%xmm2");
-			asm volatile("pxor   %xmm14,%xmm10");
-			asm volatile("pxor   %xmm7,%xmm2");
-			asm volatile("pxor   %xmm15,%xmm10");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("movdqa %xmm6,%xmm14");
+			asm volatile ("movdqa %xmm7,%xmm15");
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm12,%xmm14");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pshufb %xmm13,%xmm15");
+			asm volatile ("pxor   %xmm6,%xmm2");
+			asm volatile ("pxor   %xmm14,%xmm10");
+			asm volatile ("pxor   %xmm7,%xmm2");
+			asm volatile ("pxor   %xmm15,%xmm10");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[0][i]));
-		asm volatile("movdqa %0,%%xmm12" : : "m" (v[0][i+16]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm12" : : "m" (v[0][i+16]));
 
-		asm volatile("pxor %xmm5,%xmm5");
-		asm volatile("pxor %xmm13,%xmm13");
-		asm volatile("pcmpgtb %xmm1,%xmm5");
-		asm volatile("pcmpgtb %xmm9,%xmm13");
-		asm volatile("paddb %xmm1,%xmm1");
-		asm volatile("paddb %xmm9,%xmm9");
-		asm volatile("pand %xmm3,%xmm5");
-		asm volatile("pand %xmm3,%xmm13");
-		asm volatile("pxor %xmm5,%xmm1");
-		asm volatile("pxor %xmm13,%xmm9");
+		asm volatile ("pxor %xmm5,%xmm5");
+		asm volatile ("pxor %xmm13,%xmm13");
+		asm volatile ("pcmpgtb %xmm1,%xmm5");
+		asm volatile ("pcmpgtb %xmm9,%xmm13");
+		asm volatile ("paddb %xmm1,%xmm1");
+		asm volatile ("paddb %xmm9,%xmm9");
+		asm volatile ("pand %xmm3,%xmm5");
+		asm volatile ("pand %xmm3,%xmm13");
+		asm volatile ("pxor %xmm5,%xmm1");
+		asm volatile ("pxor %xmm13,%xmm9");
 
-		asm volatile("pxor %xmm4,%xmm0");
-		asm volatile("pxor %xmm4,%xmm1");
-		asm volatile("pxor %xmm4,%xmm2");
-		asm volatile("pxor %xmm12,%xmm8");
-		asm volatile("pxor %xmm12,%xmm9");
-		asm volatile("pxor %xmm12,%xmm10");
+		asm volatile ("pxor %xmm4,%xmm0");
+		asm volatile ("pxor %xmm4,%xmm1");
+		asm volatile ("pxor %xmm4,%xmm2");
+		asm volatile ("pxor %xmm12,%xmm8");
+		asm volatile ("pxor %xmm12,%xmm9");
+		asm volatile ("pxor %xmm12,%xmm10");
 
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm8,%0" : "=m" (p[i+16]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm9,%0" : "=m" (q[i+16]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm10,%0" : "=m" (r[i+16]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm8,%0" : "=m" (p[i+16]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm9,%0" : "=m" (q[i+16]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm10,%0" : "=m" (r[i+16]));
 	}
 
 	raid_asm_clobber_xmm16();
@@ -576,103 +576,103 @@ void raid_gen3_avx2ext(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("vbroadcasti128 %0, %%ymm3" : : "m" (gfconst16.poly[0]));
-	asm volatile("vbroadcasti128 %0, %%ymm11" : : "m" (gfconst16.low4[0]));
+	asm volatile ("vbroadcasti128 %0, %%ymm3" : : "m" (gfconst16.poly[0]));
+	asm volatile ("vbroadcasti128 %0, %%ymm11" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 64) {
 		/* last disk without the by two multiplication */
-		asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[l][i]));
-		asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[l][i+32]));
+		asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[l][i+32]));
 
-		asm volatile("vmovdqa %ymm4,%ymm0");
-		asm volatile("vmovdqa %ymm4,%ymm1");
-		asm volatile("vmovdqa %ymm12,%ymm8");
-		asm volatile("vmovdqa %ymm12,%ymm9");
+		asm volatile ("vmovdqa %ymm4,%ymm0");
+		asm volatile ("vmovdqa %ymm4,%ymm1");
+		asm volatile ("vmovdqa %ymm12,%ymm8");
+		asm volatile ("vmovdqa %ymm12,%ymm9");
 
-		asm volatile("vpsrlw  $4,%ymm4,%ymm5");
-		asm volatile("vpsrlw  $4,%ymm12,%ymm13");
-		asm volatile("vpand   %ymm11,%ymm4,%ymm4");
-		asm volatile("vpand   %ymm11,%ymm12,%ymm12");
-		asm volatile("vpand   %ymm11,%ymm5,%ymm5");
-		asm volatile("vpand   %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpsrlw  $4,%ymm4,%ymm5");
+		asm volatile ("vpsrlw  $4,%ymm12,%ymm13");
+		asm volatile ("vpand   %ymm11,%ymm4,%ymm4");
+		asm volatile ("vpand   %ymm11,%ymm12,%ymm12");
+		asm volatile ("vpand   %ymm11,%ymm5,%ymm5");
+		asm volatile ("vpand   %ymm11,%ymm13,%ymm13");
 
-		asm volatile("vbroadcasti128 %0,%%ymm10" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("vpshufb %ymm4,%ymm10,%ymm2");
-		asm volatile("vpshufb %ymm12,%ymm10,%ymm10");
-		asm volatile("vpshufb %ymm5,%ymm15,%ymm7");
-		asm volatile("vpshufb %ymm13,%ymm15,%ymm15");
-		asm volatile("vpxor   %ymm7,%ymm2,%ymm2");
-		asm volatile("vpxor   %ymm15,%ymm10,%ymm10");
+		asm volatile ("vbroadcasti128 %0,%%ymm10" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("vpshufb %ymm4,%ymm10,%ymm2");
+		asm volatile ("vpshufb %ymm12,%ymm10,%ymm10");
+		asm volatile ("vpshufb %ymm5,%ymm15,%ymm7");
+		asm volatile ("vpshufb %ymm13,%ymm15,%ymm15");
+		asm volatile ("vpxor   %ymm7,%ymm2,%ymm2");
+		asm volatile ("vpxor   %ymm15,%ymm10,%ymm10");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
-			asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[d][i+32]));
+			asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
+			asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[d][i+32]));
 
-			asm volatile("vpxor %ymm5,%ymm5,%ymm5");
-			asm volatile("vpxor %ymm13,%ymm13,%ymm13");
-			asm volatile("vpcmpgtb %ymm1,%ymm5,%ymm5");
-			asm volatile("vpcmpgtb %ymm9,%ymm13,%ymm13");
-			asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-			asm volatile("vpaddb %ymm9,%ymm9,%ymm9");
-			asm volatile("vpand %ymm3,%ymm5,%ymm5");
-			asm volatile("vpand %ymm3,%ymm13,%ymm13");
-			asm volatile("vpxor %ymm5,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm13,%ymm9,%ymm9");
+			asm volatile ("vpxor %ymm5,%ymm5,%ymm5");
+			asm volatile ("vpxor %ymm13,%ymm13,%ymm13");
+			asm volatile ("vpcmpgtb %ymm1,%ymm5,%ymm5");
+			asm volatile ("vpcmpgtb %ymm9,%ymm13,%ymm13");
+			asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+			asm volatile ("vpaddb %ymm9,%ymm9,%ymm9");
+			asm volatile ("vpand %ymm3,%ymm5,%ymm5");
+			asm volatile ("vpand %ymm3,%ymm13,%ymm13");
+			asm volatile ("vpxor %ymm5,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm13,%ymm9,%ymm9");
 
-			asm volatile("vpxor %ymm4,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm4,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm12,%ymm8,%ymm8");
-			asm volatile("vpxor %ymm12,%ymm9,%ymm9");
+			asm volatile ("vpxor %ymm4,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm4,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm12,%ymm8,%ymm8");
+			asm volatile ("vpxor %ymm12,%ymm9,%ymm9");
 
-			asm volatile("vpsrlw  $4,%ymm4,%ymm5");
-			asm volatile("vpsrlw  $4,%ymm12,%ymm13");
-			asm volatile("vpand   %ymm11,%ymm4,%ymm4");
-			asm volatile("vpand   %ymm11,%ymm12,%ymm12");
-			asm volatile("vpand   %ymm11,%ymm5,%ymm5");
-			asm volatile("vpand   %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpsrlw  $4,%ymm4,%ymm5");
+			asm volatile ("vpsrlw  $4,%ymm12,%ymm13");
+			asm volatile ("vpand   %ymm11,%ymm4,%ymm4");
+			asm volatile ("vpand   %ymm11,%ymm12,%ymm12");
+			asm volatile ("vpand   %ymm11,%ymm5,%ymm5");
+			asm volatile ("vpand   %ymm11,%ymm13,%ymm13");
 
-			asm volatile("vbroadcasti128 %0,%%ymm14" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("vpshufb %ymm4,%ymm14,%ymm6");
-			asm volatile("vpshufb %ymm12,%ymm14,%ymm14");
-			asm volatile("vpshufb %ymm5,%ymm15,%ymm7");
-			asm volatile("vpshufb %ymm13,%ymm15,%ymm15");
-			asm volatile("vpxor   %ymm6,%ymm2,%ymm2");
-			asm volatile("vpxor   %ymm14,%ymm10,%ymm10");
-			asm volatile("vpxor   %ymm7,%ymm2,%ymm2");
-			asm volatile("vpxor   %ymm15,%ymm10,%ymm10");
+			asm volatile ("vbroadcasti128 %0,%%ymm14" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("vpshufb %ymm4,%ymm14,%ymm6");
+			asm volatile ("vpshufb %ymm12,%ymm14,%ymm14");
+			asm volatile ("vpshufb %ymm5,%ymm15,%ymm7");
+			asm volatile ("vpshufb %ymm13,%ymm15,%ymm15");
+			asm volatile ("vpxor   %ymm6,%ymm2,%ymm2");
+			asm volatile ("vpxor   %ymm14,%ymm10,%ymm10");
+			asm volatile ("vpxor   %ymm7,%ymm2,%ymm2");
+			asm volatile ("vpxor   %ymm15,%ymm10,%ymm10");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[0][i]));
-		asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[0][i+32]));
+		asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[0][i]));
+		asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[0][i+32]));
 
-		asm volatile("vpxor %ymm5,%ymm5,%ymm5");
-		asm volatile("vpxor %ymm13,%ymm13,%ymm13");
-		asm volatile("vpcmpgtb %ymm1,%ymm5,%ymm5");
-		asm volatile("vpcmpgtb %ymm9,%ymm13,%ymm13");
-		asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-		asm volatile("vpaddb %ymm9,%ymm9,%ymm9");
-		asm volatile("vpand %ymm3,%ymm5,%ymm5");
-		asm volatile("vpand %ymm3,%ymm13,%ymm13");
-		asm volatile("vpxor %ymm5,%ymm1,%ymm1");
-		asm volatile("vpxor %ymm13,%ymm9,%ymm9");
+		asm volatile ("vpxor %ymm5,%ymm5,%ymm5");
+		asm volatile ("vpxor %ymm13,%ymm13,%ymm13");
+		asm volatile ("vpcmpgtb %ymm1,%ymm5,%ymm5");
+		asm volatile ("vpcmpgtb %ymm9,%ymm13,%ymm13");
+		asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+		asm volatile ("vpaddb %ymm9,%ymm9,%ymm9");
+		asm volatile ("vpand %ymm3,%ymm5,%ymm5");
+		asm volatile ("vpand %ymm3,%ymm13,%ymm13");
+		asm volatile ("vpxor %ymm5,%ymm1,%ymm1");
+		asm volatile ("vpxor %ymm13,%ymm9,%ymm9");
 
-		asm volatile("vpxor %ymm4,%ymm0,%ymm0");
-		asm volatile("vpxor %ymm4,%ymm1,%ymm1");
-		asm volatile("vpxor %ymm4,%ymm2,%ymm2");
-		asm volatile("vpxor %ymm12,%ymm8,%ymm8");
-		asm volatile("vpxor %ymm12,%ymm9,%ymm9");
-		asm volatile("vpxor %ymm12,%ymm10,%ymm10");
+		asm volatile ("vpxor %ymm4,%ymm0,%ymm0");
+		asm volatile ("vpxor %ymm4,%ymm1,%ymm1");
+		asm volatile ("vpxor %ymm4,%ymm2,%ymm2");
+		asm volatile ("vpxor %ymm12,%ymm8,%ymm8");
+		asm volatile ("vpxor %ymm12,%ymm9,%ymm9");
+		asm volatile ("vpxor %ymm12,%ymm10,%ymm10");
 
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm8,%0" : "=m" (p[i+32]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (q[i]));
-		asm volatile("vmovntdq %%ymm9,%0" : "=m" (q[i+32]));
-		asm volatile("vmovntdq %%ymm2,%0" : "=m" (r[i]));
-		asm volatile("vmovntdq %%ymm10,%0" : "=m" (r[i+32]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm8,%0" : "=m" (p[i+32]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (q[i]));
+		asm volatile ("vmovntdq %%ymm9,%0" : "=m" (q[i+32]));
+		asm volatile ("vmovntdq %%ymm2,%0" : "=m" (r[i]));
+		asm volatile ("vmovntdq %%ymm10,%0" : "=m" (r[i+32]));
 	}
 
 	raid_asm_clobber_ymm16();
@@ -713,84 +713,84 @@ void raid_gen4_ssse3(int nd, size_t size, void **vv)
 	/* generic case with at least two data disks */
 	for (i = 0; i < size; i += 16) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[l][i]));
 
-		asm volatile("movdqa %xmm4,%xmm0");
-		asm volatile("movdqa %xmm4,%xmm1");
+		asm volatile ("movdqa %xmm4,%xmm0");
+		asm volatile ("movdqa %xmm4,%xmm1");
 
-		asm volatile("movdqa %xmm4,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
+		asm volatile ("movdqa %xmm4,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm2");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm2");
 
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("pshufb %xmm4,%xmm3");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm3");
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm3");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm3");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
 
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pcmpgtb %xmm1,%xmm5");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("pand %xmm7,%xmm5");
-			asm volatile("pxor %xmm5,%xmm1");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pcmpgtb %xmm1,%xmm5");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("pand %xmm7,%xmm5");
+			asm volatile ("pxor %xmm5,%xmm1");
 
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
 
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm1");
 
-			asm volatile("movdqa %xmm4,%xmm5");
-			asm volatile("psrlw  $4,%xmm5");
-			asm volatile("pand   %xmm7,%xmm4");
-			asm volatile("pand   %xmm7,%xmm5");
+			asm volatile ("movdqa %xmm4,%xmm5");
+			asm volatile ("psrlw  $4,%xmm5");
+			asm volatile ("pand   %xmm7,%xmm4");
+			asm volatile ("pand   %xmm7,%xmm5");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm2");
-			asm volatile("pxor   %xmm7,%xmm2");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm2");
+			asm volatile ("pxor   %xmm7,%xmm2");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm3");
-			asm volatile("pxor   %xmm7,%xmm3");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm3");
+			asm volatile ("pxor   %xmm7,%xmm3");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[0][i]));
 
-		asm volatile("pxor %xmm5,%xmm5");
-		asm volatile("pcmpgtb %xmm1,%xmm5");
-		asm volatile("paddb %xmm1,%xmm1");
-		asm volatile("pand %xmm7,%xmm5");
-		asm volatile("pxor %xmm5,%xmm1");
+		asm volatile ("pxor %xmm5,%xmm5");
+		asm volatile ("pcmpgtb %xmm1,%xmm5");
+		asm volatile ("paddb %xmm1,%xmm1");
+		asm volatile ("pand %xmm7,%xmm5");
+		asm volatile ("pxor %xmm5,%xmm1");
 
-		asm volatile("pxor %xmm4,%xmm0");
-		asm volatile("pxor %xmm4,%xmm1");
-		asm volatile("pxor %xmm4,%xmm2");
-		asm volatile("pxor %xmm4,%xmm3");
+		asm volatile ("pxor %xmm4,%xmm0");
+		asm volatile ("pxor %xmm4,%xmm1");
+		asm volatile ("pxor %xmm4,%xmm2");
+		asm volatile ("pxor %xmm4,%xmm3");
 
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (s[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (s[i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -833,139 +833,139 @@ void raid_gen4_ssse3ext(int nd, size_t size, void **vv)
 	/* generic case with at least two data disks */
 	for (i = 0; i < size; i += 32) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[l][i]));
-		asm volatile("movdqa %0,%%xmm12" : : "m" (v[l][i+16]));
+		asm volatile ("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm12" : : "m" (v[l][i+16]));
 
-		asm volatile("movdqa %xmm4,%xmm0");
-		asm volatile("movdqa %xmm4,%xmm1");
-		asm volatile("movdqa %xmm12,%xmm8");
-		asm volatile("movdqa %xmm12,%xmm9");
+		asm volatile ("movdqa %xmm4,%xmm0");
+		asm volatile ("movdqa %xmm4,%xmm1");
+		asm volatile ("movdqa %xmm12,%xmm8");
+		asm volatile ("movdqa %xmm12,%xmm9");
 
-		asm volatile("movdqa %xmm4,%xmm5");
-		asm volatile("movdqa %xmm12,%xmm13");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("psrlw  $4,%xmm13");
-		asm volatile("pand   %xmm15,%xmm4");
-		asm volatile("pand   %xmm15,%xmm12");
-		asm volatile("pand   %xmm15,%xmm5");
-		asm volatile("pand   %xmm15,%xmm13");
+		asm volatile ("movdqa %xmm4,%xmm5");
+		asm volatile ("movdqa %xmm12,%xmm13");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("psrlw  $4,%xmm13");
+		asm volatile ("pand   %xmm15,%xmm4");
+		asm volatile ("pand   %xmm15,%xmm12");
+		asm volatile ("pand   %xmm15,%xmm5");
+		asm volatile ("pand   %xmm15,%xmm13");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("movdqa %xmm2,%xmm10");
-		asm volatile("movdqa %xmm7,%xmm15");
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm12,%xmm10");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pshufb %xmm13,%xmm15");
-		asm volatile("pxor   %xmm7,%xmm2");
-		asm volatile("pxor   %xmm15,%xmm10");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("movdqa %xmm2,%xmm10");
+		asm volatile ("movdqa %xmm7,%xmm15");
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm12,%xmm10");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pshufb %xmm13,%xmm15");
+		asm volatile ("pxor   %xmm7,%xmm2");
+		asm volatile ("pxor   %xmm15,%xmm10");
 
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("movdqa %xmm3,%xmm11");
-		asm volatile("movdqa %xmm7,%xmm15");
-		asm volatile("pshufb %xmm4,%xmm3");
-		asm volatile("pshufb %xmm12,%xmm11");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pshufb %xmm13,%xmm15");
-		asm volatile("pxor   %xmm7,%xmm3");
-		asm volatile("pxor   %xmm15,%xmm11");
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("movdqa %xmm3,%xmm11");
+		asm volatile ("movdqa %xmm7,%xmm15");
+		asm volatile ("pshufb %xmm4,%xmm3");
+		asm volatile ("pshufb %xmm12,%xmm11");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pshufb %xmm13,%xmm15");
+		asm volatile ("pxor   %xmm7,%xmm3");
+		asm volatile ("pxor   %xmm15,%xmm11");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
-			asm volatile("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
-			asm volatile("movdqa %0,%%xmm12" : : "m" (v[d][i+16]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+			asm volatile ("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (v[d][i+16]));
 
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pxor %xmm13,%xmm13");
-			asm volatile("pcmpgtb %xmm1,%xmm5");
-			asm volatile("pcmpgtb %xmm9,%xmm13");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("paddb %xmm9,%xmm9");
-			asm volatile("pand %xmm7,%xmm5");
-			asm volatile("pand %xmm7,%xmm13");
-			asm volatile("pxor %xmm5,%xmm1");
-			asm volatile("pxor %xmm13,%xmm9");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pxor %xmm13,%xmm13");
+			asm volatile ("pcmpgtb %xmm1,%xmm5");
+			asm volatile ("pcmpgtb %xmm9,%xmm13");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("paddb %xmm9,%xmm9");
+			asm volatile ("pand %xmm7,%xmm5");
+			asm volatile ("pand %xmm7,%xmm13");
+			asm volatile ("pxor %xmm5,%xmm1");
+			asm volatile ("pxor %xmm13,%xmm9");
 
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm1");
-			asm volatile("pxor %xmm12,%xmm8");
-			asm volatile("pxor %xmm12,%xmm9");
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm12,%xmm8");
+			asm volatile ("pxor %xmm12,%xmm9");
 
-			asm volatile("movdqa %xmm4,%xmm5");
-			asm volatile("movdqa %xmm12,%xmm13");
-			asm volatile("psrlw  $4,%xmm5");
-			asm volatile("psrlw  $4,%xmm13");
-			asm volatile("pand   %xmm15,%xmm4");
-			asm volatile("pand   %xmm15,%xmm12");
-			asm volatile("pand   %xmm15,%xmm5");
-			asm volatile("pand   %xmm15,%xmm13");
+			asm volatile ("movdqa %xmm4,%xmm5");
+			asm volatile ("movdqa %xmm12,%xmm13");
+			asm volatile ("psrlw  $4,%xmm5");
+			asm volatile ("psrlw  $4,%xmm13");
+			asm volatile ("pand   %xmm15,%xmm4");
+			asm volatile ("pand   %xmm15,%xmm12");
+			asm volatile ("pand   %xmm15,%xmm5");
+			asm volatile ("pand   %xmm15,%xmm13");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("movdqa %xmm6,%xmm14");
-			asm volatile("movdqa %xmm7,%xmm15");
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm12,%xmm14");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pshufb %xmm13,%xmm15");
-			asm volatile("pxor   %xmm6,%xmm2");
-			asm volatile("pxor   %xmm14,%xmm10");
-			asm volatile("pxor   %xmm7,%xmm2");
-			asm volatile("pxor   %xmm15,%xmm10");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("movdqa %xmm6,%xmm14");
+			asm volatile ("movdqa %xmm7,%xmm15");
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm12,%xmm14");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pshufb %xmm13,%xmm15");
+			asm volatile ("pxor   %xmm6,%xmm2");
+			asm volatile ("pxor   %xmm14,%xmm10");
+			asm volatile ("pxor   %xmm7,%xmm2");
+			asm volatile ("pxor   %xmm15,%xmm10");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("movdqa %xmm6,%xmm14");
-			asm volatile("movdqa %xmm7,%xmm15");
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm12,%xmm14");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pshufb %xmm13,%xmm15");
-			asm volatile("pxor   %xmm6,%xmm3");
-			asm volatile("pxor   %xmm14,%xmm11");
-			asm volatile("pxor   %xmm7,%xmm3");
-			asm volatile("pxor   %xmm15,%xmm11");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("movdqa %xmm6,%xmm14");
+			asm volatile ("movdqa %xmm7,%xmm15");
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm12,%xmm14");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pshufb %xmm13,%xmm15");
+			asm volatile ("pxor   %xmm6,%xmm3");
+			asm volatile ("pxor   %xmm14,%xmm11");
+			asm volatile ("pxor   %xmm7,%xmm3");
+			asm volatile ("pxor   %xmm15,%xmm11");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
-		asm volatile("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[0][i]));
-		asm volatile("movdqa %0,%%xmm12" : : "m" (v[0][i+16]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+		asm volatile ("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm12" : : "m" (v[0][i+16]));
 
-		asm volatile("pxor %xmm5,%xmm5");
-		asm volatile("pxor %xmm13,%xmm13");
-		asm volatile("pcmpgtb %xmm1,%xmm5");
-		asm volatile("pcmpgtb %xmm9,%xmm13");
-		asm volatile("paddb %xmm1,%xmm1");
-		asm volatile("paddb %xmm9,%xmm9");
-		asm volatile("pand %xmm7,%xmm5");
-		asm volatile("pand %xmm7,%xmm13");
-		asm volatile("pxor %xmm5,%xmm1");
-		asm volatile("pxor %xmm13,%xmm9");
+		asm volatile ("pxor %xmm5,%xmm5");
+		asm volatile ("pxor %xmm13,%xmm13");
+		asm volatile ("pcmpgtb %xmm1,%xmm5");
+		asm volatile ("pcmpgtb %xmm9,%xmm13");
+		asm volatile ("paddb %xmm1,%xmm1");
+		asm volatile ("paddb %xmm9,%xmm9");
+		asm volatile ("pand %xmm7,%xmm5");
+		asm volatile ("pand %xmm7,%xmm13");
+		asm volatile ("pxor %xmm5,%xmm1");
+		asm volatile ("pxor %xmm13,%xmm9");
 
-		asm volatile("pxor %xmm4,%xmm0");
-		asm volatile("pxor %xmm4,%xmm1");
-		asm volatile("pxor %xmm4,%xmm2");
-		asm volatile("pxor %xmm4,%xmm3");
-		asm volatile("pxor %xmm12,%xmm8");
-		asm volatile("pxor %xmm12,%xmm9");
-		asm volatile("pxor %xmm12,%xmm10");
-		asm volatile("pxor %xmm12,%xmm11");
+		asm volatile ("pxor %xmm4,%xmm0");
+		asm volatile ("pxor %xmm4,%xmm1");
+		asm volatile ("pxor %xmm4,%xmm2");
+		asm volatile ("pxor %xmm4,%xmm3");
+		asm volatile ("pxor %xmm12,%xmm8");
+		asm volatile ("pxor %xmm12,%xmm9");
+		asm volatile ("pxor %xmm12,%xmm10");
+		asm volatile ("pxor %xmm12,%xmm11");
 
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm8,%0" : "=m" (p[i+16]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm9,%0" : "=m" (q[i+16]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm10,%0" : "=m" (r[i+16]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (s[i]));
-		asm volatile("movntdq %%xmm11,%0" : "=m" (s[i+16]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm8,%0" : "=m" (p[i+16]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm9,%0" : "=m" (q[i+16]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm10,%0" : "=m" (r[i+16]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (s[i]));
+		asm volatile ("movntdq %%xmm11,%0" : "=m" (s[i+16]));
 	}
 
 	raid_asm_clobber_xmm16();
@@ -1008,127 +1008,127 @@ void raid_gen4_avx2ext(int nd, size_t size, void **vv)
 	/* generic case with at least two data disks */
 	for (i = 0; i < size; i += 64) {
 		/* last disk without the by two multiplication */
-		asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
-		asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[l][i]));
-		asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[l][i+32]));
+		asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
+		asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[l][i+32]));
 
-		asm volatile("vmovdqa %ymm4,%ymm0");
-		asm volatile("vmovdqa %ymm4,%ymm1");
-		asm volatile("vmovdqa %ymm12,%ymm8");
-		asm volatile("vmovdqa %ymm12,%ymm9");
+		asm volatile ("vmovdqa %ymm4,%ymm0");
+		asm volatile ("vmovdqa %ymm4,%ymm1");
+		asm volatile ("vmovdqa %ymm12,%ymm8");
+		asm volatile ("vmovdqa %ymm12,%ymm9");
 
-		asm volatile("vpsrlw  $4,%ymm4,%ymm5");
-		asm volatile("vpsrlw  $4,%ymm12,%ymm13");
-		asm volatile("vpand   %ymm15,%ymm4,%ymm4");
-		asm volatile("vpand   %ymm15,%ymm12,%ymm12");
-		asm volatile("vpand   %ymm15,%ymm5,%ymm5");
-		asm volatile("vpand   %ymm15,%ymm13,%ymm13");
+		asm volatile ("vpsrlw  $4,%ymm4,%ymm5");
+		asm volatile ("vpsrlw  $4,%ymm12,%ymm13");
+		asm volatile ("vpand   %ymm15,%ymm4,%ymm4");
+		asm volatile ("vpand   %ymm15,%ymm12,%ymm12");
+		asm volatile ("vpand   %ymm15,%ymm5,%ymm5");
+		asm volatile ("vpand   %ymm15,%ymm13,%ymm13");
 
-		asm volatile("vbroadcasti128 %0,%%ymm10" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("vpshufb %ymm4,%ymm10,%ymm2");
-		asm volatile("vpshufb %ymm5,%ymm15,%ymm7");
-		asm volatile("vpshufb %ymm12,%ymm10,%ymm10");
-		asm volatile("vpshufb %ymm13,%ymm15,%ymm15");
-		asm volatile("vpxor   %ymm7,%ymm2,%ymm2");
-		asm volatile("vpxor   %ymm15,%ymm10,%ymm10");
+		asm volatile ("vbroadcasti128 %0,%%ymm10" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("vpshufb %ymm4,%ymm10,%ymm2");
+		asm volatile ("vpshufb %ymm5,%ymm15,%ymm7");
+		asm volatile ("vpshufb %ymm12,%ymm10,%ymm10");
+		asm volatile ("vpshufb %ymm13,%ymm15,%ymm15");
+		asm volatile ("vpxor   %ymm7,%ymm2,%ymm2");
+		asm volatile ("vpxor   %ymm15,%ymm10,%ymm10");
 
-		asm volatile("vbroadcasti128 %0,%%ymm11" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("vpshufb %ymm4,%ymm11,%ymm3");
-		asm volatile("vpshufb %ymm5,%ymm15,%ymm7");
-		asm volatile("vpshufb %ymm12,%ymm11,%ymm11");
-		asm volatile("vpshufb %ymm13,%ymm15,%ymm15");
-		asm volatile("vpxor   %ymm7,%ymm3,%ymm3");
-		asm volatile("vpxor   %ymm15,%ymm11,%ymm11");
+		asm volatile ("vbroadcasti128 %0,%%ymm11" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("vpshufb %ymm4,%ymm11,%ymm3");
+		asm volatile ("vpshufb %ymm5,%ymm15,%ymm7");
+		asm volatile ("vpshufb %ymm12,%ymm11,%ymm11");
+		asm volatile ("vpshufb %ymm13,%ymm15,%ymm15");
+		asm volatile ("vpxor   %ymm7,%ymm3,%ymm3");
+		asm volatile ("vpxor   %ymm15,%ymm11,%ymm11");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.poly[0]));
-			asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
-			asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
-			asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[d][i+32]));
+			asm volatile ("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.poly[0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
+			asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
+			asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[d][i+32]));
 
-			asm volatile("vpxor %ymm5,%ymm5,%ymm5");
-			asm volatile("vpxor %ymm13,%ymm13,%ymm13");
-			asm volatile("vpcmpgtb %ymm1,%ymm5,%ymm5");
-			asm volatile("vpcmpgtb %ymm9,%ymm13,%ymm13");
-			asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-			asm volatile("vpaddb %ymm9,%ymm9,%ymm9");
-			asm volatile("vpand %ymm7,%ymm5,%ymm5");
-			asm volatile("vpand %ymm7,%ymm13,%ymm13");
-			asm volatile("vpxor %ymm5,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm13,%ymm9,%ymm9");
+			asm volatile ("vpxor %ymm5,%ymm5,%ymm5");
+			asm volatile ("vpxor %ymm13,%ymm13,%ymm13");
+			asm volatile ("vpcmpgtb %ymm1,%ymm5,%ymm5");
+			asm volatile ("vpcmpgtb %ymm9,%ymm13,%ymm13");
+			asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+			asm volatile ("vpaddb %ymm9,%ymm9,%ymm9");
+			asm volatile ("vpand %ymm7,%ymm5,%ymm5");
+			asm volatile ("vpand %ymm7,%ymm13,%ymm13");
+			asm volatile ("vpxor %ymm5,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm13,%ymm9,%ymm9");
 
-			asm volatile("vpxor %ymm4,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm4,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm12,%ymm8,%ymm8");
-			asm volatile("vpxor %ymm12,%ymm9,%ymm9");
+			asm volatile ("vpxor %ymm4,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm4,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm12,%ymm8,%ymm8");
+			asm volatile ("vpxor %ymm12,%ymm9,%ymm9");
 
-			asm volatile("vpsrlw  $4,%ymm4,%ymm5");
-			asm volatile("vpsrlw  $4,%ymm12,%ymm13");
-			asm volatile("vpand   %ymm15,%ymm4,%ymm4");
-			asm volatile("vpand   %ymm15,%ymm12,%ymm12");
-			asm volatile("vpand   %ymm15,%ymm5,%ymm5");
-			asm volatile("vpand   %ymm15,%ymm13,%ymm13");
+			asm volatile ("vpsrlw  $4,%ymm4,%ymm5");
+			asm volatile ("vpsrlw  $4,%ymm12,%ymm13");
+			asm volatile ("vpand   %ymm15,%ymm4,%ymm4");
+			asm volatile ("vpand   %ymm15,%ymm12,%ymm12");
+			asm volatile ("vpand   %ymm15,%ymm5,%ymm5");
+			asm volatile ("vpand   %ymm15,%ymm13,%ymm13");
 
-			asm volatile("vbroadcasti128 %0,%%ymm14" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("vpshufb %ymm4,%ymm14,%ymm6");
-			asm volatile("vpshufb %ymm5,%ymm15,%ymm7");
-			asm volatile("vpshufb %ymm12,%ymm14,%ymm14");
-			asm volatile("vpshufb %ymm13,%ymm15,%ymm15");
-			asm volatile("vpxor   %ymm6,%ymm2,%ymm2");
-			asm volatile("vpxor   %ymm14,%ymm10,%ymm10");
-			asm volatile("vpxor   %ymm7,%ymm2,%ymm2");
-			asm volatile("vpxor   %ymm15,%ymm10,%ymm10");
+			asm volatile ("vbroadcasti128 %0,%%ymm14" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("vpshufb %ymm4,%ymm14,%ymm6");
+			asm volatile ("vpshufb %ymm5,%ymm15,%ymm7");
+			asm volatile ("vpshufb %ymm12,%ymm14,%ymm14");
+			asm volatile ("vpshufb %ymm13,%ymm15,%ymm15");
+			asm volatile ("vpxor   %ymm6,%ymm2,%ymm2");
+			asm volatile ("vpxor   %ymm14,%ymm10,%ymm10");
+			asm volatile ("vpxor   %ymm7,%ymm2,%ymm2");
+			asm volatile ("vpxor   %ymm15,%ymm10,%ymm10");
 
-			asm volatile("vbroadcasti128 %0,%%ymm14" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("vpshufb %ymm4,%ymm14,%ymm6");
-			asm volatile("vpshufb %ymm5,%ymm15,%ymm7");
-			asm volatile("vpshufb %ymm12,%ymm14,%ymm14");
-			asm volatile("vpshufb %ymm13,%ymm15,%ymm15");
-			asm volatile("vpxor   %ymm6,%ymm3,%ymm3");
-			asm volatile("vpxor   %ymm14,%ymm11,%ymm11");
-			asm volatile("vpxor   %ymm7,%ymm3,%ymm3");
-			asm volatile("vpxor   %ymm15,%ymm11,%ymm11");
+			asm volatile ("vbroadcasti128 %0,%%ymm14" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("vpshufb %ymm4,%ymm14,%ymm6");
+			asm volatile ("vpshufb %ymm5,%ymm15,%ymm7");
+			asm volatile ("vpshufb %ymm12,%ymm14,%ymm14");
+			asm volatile ("vpshufb %ymm13,%ymm15,%ymm15");
+			asm volatile ("vpxor   %ymm6,%ymm3,%ymm3");
+			asm volatile ("vpxor   %ymm14,%ymm11,%ymm11");
+			asm volatile ("vpxor   %ymm7,%ymm3,%ymm3");
+			asm volatile ("vpxor   %ymm15,%ymm11,%ymm11");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.poly[0]));
-		asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
-		asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[0][i]));
-		asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[0][i+32]));
+		asm volatile ("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.poly[0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
+		asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[0][i]));
+		asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[0][i+32]));
 
-		asm volatile("vpxor %ymm5,%ymm5,%ymm5");
-		asm volatile("vpxor %ymm13,%ymm13,%ymm13");
-		asm volatile("vpcmpgtb %ymm1,%ymm5,%ymm5");
-		asm volatile("vpcmpgtb %ymm9,%ymm13,%ymm13");
-		asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-		asm volatile("vpaddb %ymm9,%ymm9,%ymm9");
-		asm volatile("vpand %ymm7,%ymm5,%ymm5");
-		asm volatile("vpand %ymm7,%ymm13,%ymm13");
-		asm volatile("vpxor %ymm5,%ymm1,%ymm1");
-		asm volatile("vpxor %ymm13,%ymm9,%ymm9");
+		asm volatile ("vpxor %ymm5,%ymm5,%ymm5");
+		asm volatile ("vpxor %ymm13,%ymm13,%ymm13");
+		asm volatile ("vpcmpgtb %ymm1,%ymm5,%ymm5");
+		asm volatile ("vpcmpgtb %ymm9,%ymm13,%ymm13");
+		asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+		asm volatile ("vpaddb %ymm9,%ymm9,%ymm9");
+		asm volatile ("vpand %ymm7,%ymm5,%ymm5");
+		asm volatile ("vpand %ymm7,%ymm13,%ymm13");
+		asm volatile ("vpxor %ymm5,%ymm1,%ymm1");
+		asm volatile ("vpxor %ymm13,%ymm9,%ymm9");
 
-		asm volatile("vpxor %ymm4,%ymm0,%ymm0");
-		asm volatile("vpxor %ymm4,%ymm1,%ymm1");
-		asm volatile("vpxor %ymm4,%ymm2,%ymm2");
-		asm volatile("vpxor %ymm4,%ymm3,%ymm3");
-		asm volatile("vpxor %ymm12,%ymm8,%ymm8");
-		asm volatile("vpxor %ymm12,%ymm9,%ymm9");
-		asm volatile("vpxor %ymm12,%ymm10,%ymm10");
-		asm volatile("vpxor %ymm12,%ymm11,%ymm11");
+		asm volatile ("vpxor %ymm4,%ymm0,%ymm0");
+		asm volatile ("vpxor %ymm4,%ymm1,%ymm1");
+		asm volatile ("vpxor %ymm4,%ymm2,%ymm2");
+		asm volatile ("vpxor %ymm4,%ymm3,%ymm3");
+		asm volatile ("vpxor %ymm12,%ymm8,%ymm8");
+		asm volatile ("vpxor %ymm12,%ymm9,%ymm9");
+		asm volatile ("vpxor %ymm12,%ymm10,%ymm10");
+		asm volatile ("vpxor %ymm12,%ymm11,%ymm11");
 
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm8,%0" : "=m" (p[i+32]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (q[i]));
-		asm volatile("vmovntdq %%ymm9,%0" : "=m" (q[i+32]));
-		asm volatile("vmovntdq %%ymm2,%0" : "=m" (r[i]));
-		asm volatile("vmovntdq %%ymm10,%0" : "=m" (r[i+32]));
-		asm volatile("vmovntdq %%ymm3,%0" : "=m" (s[i]));
-		asm volatile("vmovntdq %%ymm11,%0" : "=m" (s[i+32]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm8,%0" : "=m" (p[i+32]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (q[i]));
+		asm volatile ("vmovntdq %%ymm9,%0" : "=m" (q[i+32]));
+		asm volatile ("vmovntdq %%ymm2,%0" : "=m" (r[i]));
+		asm volatile ("vmovntdq %%ymm10,%0" : "=m" (r[i+32]));
+		asm volatile ("vmovntdq %%ymm3,%0" : "=m" (s[i]));
+		asm volatile ("vmovntdq %%ymm11,%0" : "=m" (s[i+32]));
 	}
 
 	raid_asm_clobber_ymm16();
@@ -1176,101 +1176,101 @@ void raid_gen5_ssse3(int nd, size_t size, void **vv)
 	/* generic case with at least two data disks */
 	for (i = 0; i < size; i += 16) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[l][i]));
 
-		asm volatile("movdqa %xmm4,%xmm0");
-		asm volatile("movdqa %%xmm4,%0" : "=m" (p0[0]));
+		asm volatile ("movdqa %xmm4,%xmm0");
+		asm volatile ("movdqa %%xmm4,%0" : "=m" (p0[0]));
 
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
-		asm volatile("movdqa %xmm4,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+		asm volatile ("movdqa %xmm4,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
 
-		asm volatile("movdqa %0,%%xmm1" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("pshufb %xmm4,%xmm1");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm1");
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm1");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm1");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm2");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm2");
 
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][2][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][2][1][0]));
-		asm volatile("pshufb %xmm4,%xmm3");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm3");
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][2][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][2][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm3");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm3");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
-			asm volatile("movdqa %0,%%xmm6" : : "m" (p0[0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (p0[0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
 
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pcmpgtb %xmm0,%xmm5");
-			asm volatile("paddb %xmm0,%xmm0");
-			asm volatile("pand %xmm7,%xmm5");
-			asm volatile("pxor %xmm5,%xmm0");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pcmpgtb %xmm0,%xmm5");
+			asm volatile ("paddb %xmm0,%xmm0");
+			asm volatile ("pand %xmm7,%xmm5");
+			asm volatile ("pxor %xmm5,%xmm0");
 
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm6");
-			asm volatile("movdqa %%xmm6,%0" : "=m" (p0[0]));
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm6");
+			asm volatile ("movdqa %%xmm6,%0" : "=m" (p0[0]));
 
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
-			asm volatile("movdqa %xmm4,%xmm5");
-			asm volatile("psrlw  $4,%xmm5");
-			asm volatile("pand   %xmm7,%xmm4");
-			asm volatile("pand   %xmm7,%xmm5");
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+			asm volatile ("movdqa %xmm4,%xmm5");
+			asm volatile ("psrlw  $4,%xmm5");
+			asm volatile ("pand   %xmm7,%xmm4");
+			asm volatile ("pand   %xmm7,%xmm5");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm1");
-			asm volatile("pxor   %xmm7,%xmm1");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm1");
+			asm volatile ("pxor   %xmm7,%xmm1");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm2");
-			asm volatile("pxor   %xmm7,%xmm2");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm2");
+			asm volatile ("pxor   %xmm7,%xmm2");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][2][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][2][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm3");
-			asm volatile("pxor   %xmm7,%xmm3");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][2][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][2][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm3");
+			asm volatile ("pxor   %xmm7,%xmm3");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[0][i]));
-		asm volatile("movdqa %0,%%xmm6" : : "m" (p0[0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm6" : : "m" (p0[0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
 
-		asm volatile("pxor %xmm5,%xmm5");
-		asm volatile("pcmpgtb %xmm0,%xmm5");
-		asm volatile("paddb %xmm0,%xmm0");
-		asm volatile("pand %xmm7,%xmm5");
-		asm volatile("pxor %xmm5,%xmm0");
+		asm volatile ("pxor %xmm5,%xmm5");
+		asm volatile ("pcmpgtb %xmm0,%xmm5");
+		asm volatile ("paddb %xmm0,%xmm0");
+		asm volatile ("pand %xmm7,%xmm5");
+		asm volatile ("pxor %xmm5,%xmm0");
 
-		asm volatile("pxor %xmm4,%xmm0");
-		asm volatile("pxor %xmm4,%xmm1");
-		asm volatile("pxor %xmm4,%xmm2");
-		asm volatile("pxor %xmm4,%xmm3");
-		asm volatile("pxor %xmm4,%xmm6");
+		asm volatile ("pxor %xmm4,%xmm0");
+		asm volatile ("pxor %xmm4,%xmm1");
+		asm volatile ("pxor %xmm4,%xmm2");
+		asm volatile ("pxor %xmm4,%xmm3");
+		asm volatile ("pxor %xmm4,%xmm6");
 
-		asm volatile("movntdq %%xmm6,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm0,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (s[i]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (t[i]));
+		asm volatile ("movntdq %%xmm6,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (s[i]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (t[i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -1313,99 +1313,99 @@ void raid_gen5_ssse3ext(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("movdqa %0,%%xmm14" : : "m" (gfconst16.poly[0]));
-	asm volatile("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm14" : : "m" (gfconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 16) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm10" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm10" : : "m" (v[l][i]));
 
-		asm volatile("movdqa %xmm10,%xmm0");
-		asm volatile("movdqa %xmm10,%xmm1");
+		asm volatile ("movdqa %xmm10,%xmm0");
+		asm volatile ("movdqa %xmm10,%xmm1");
 
-		asm volatile("movdqa %xmm10,%xmm11");
-		asm volatile("psrlw  $4,%xmm11");
-		asm volatile("pand   %xmm15,%xmm10");
-		asm volatile("pand   %xmm15,%xmm11");
+		asm volatile ("movdqa %xmm10,%xmm11");
+		asm volatile ("psrlw  $4,%xmm11");
+		asm volatile ("pand   %xmm15,%xmm10");
+		asm volatile ("pand   %xmm15,%xmm11");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("pshufb %xmm10,%xmm2");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm2");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm2");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm2");
 
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("pshufb %xmm10,%xmm3");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm3");
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm3");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm3");
 
-		asm volatile("movdqa %0,%%xmm4" : : "m" (gfgenpshufb[l][2][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][2][1][0]));
-		asm volatile("pshufb %xmm10,%xmm4");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm4");
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (gfgenpshufb[l][2][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][2][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm4");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm4");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm10" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm10" : : "m" (v[d][i]));
 
-			asm volatile("pxor %xmm11,%xmm11");
-			asm volatile("pcmpgtb %xmm1,%xmm11");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("pand %xmm14,%xmm11");
-			asm volatile("pxor %xmm11,%xmm1");
+			asm volatile ("pxor %xmm11,%xmm11");
+			asm volatile ("pcmpgtb %xmm1,%xmm11");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("pand %xmm14,%xmm11");
+			asm volatile ("pxor %xmm11,%xmm1");
 
-			asm volatile("pxor %xmm10,%xmm0");
-			asm volatile("pxor %xmm10,%xmm1");
+			asm volatile ("pxor %xmm10,%xmm0");
+			asm volatile ("pxor %xmm10,%xmm1");
 
-			asm volatile("movdqa %xmm10,%xmm11");
-			asm volatile("psrlw  $4,%xmm11");
-			asm volatile("pand   %xmm15,%xmm10");
-			asm volatile("pand   %xmm15,%xmm11");
+			asm volatile ("movdqa %xmm10,%xmm11");
+			asm volatile ("psrlw  $4,%xmm11");
+			asm volatile ("pand   %xmm15,%xmm10");
+			asm volatile ("pand   %xmm15,%xmm11");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm2");
-			asm volatile("pxor   %xmm13,%xmm2");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm2");
+			asm volatile ("pxor   %xmm13,%xmm2");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm3");
-			asm volatile("pxor   %xmm13,%xmm3");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm3");
+			asm volatile ("pxor   %xmm13,%xmm3");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][2][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][2][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm4");
-			asm volatile("pxor   %xmm13,%xmm4");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][2][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][2][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm4");
+			asm volatile ("pxor   %xmm13,%xmm4");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm10" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm10" : : "m" (v[0][i]));
 
-		asm volatile("pxor %xmm11,%xmm11");
-		asm volatile("pcmpgtb %xmm1,%xmm11");
-		asm volatile("paddb %xmm1,%xmm1");
-		asm volatile("pand %xmm14,%xmm11");
-		asm volatile("pxor %xmm11,%xmm1");
+		asm volatile ("pxor %xmm11,%xmm11");
+		asm volatile ("pcmpgtb %xmm1,%xmm11");
+		asm volatile ("paddb %xmm1,%xmm1");
+		asm volatile ("pand %xmm14,%xmm11");
+		asm volatile ("pxor %xmm11,%xmm1");
 
-		asm volatile("pxor %xmm10,%xmm0");
-		asm volatile("pxor %xmm10,%xmm1");
-		asm volatile("pxor %xmm10,%xmm2");
-		asm volatile("pxor %xmm10,%xmm3");
-		asm volatile("pxor %xmm10,%xmm4");
+		asm volatile ("pxor %xmm10,%xmm0");
+		asm volatile ("pxor %xmm10,%xmm1");
+		asm volatile ("pxor %xmm10,%xmm2");
+		asm volatile ("pxor %xmm10,%xmm3");
+		asm volatile ("pxor %xmm10,%xmm4");
 
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (s[i]));
-		asm volatile("movntdq %%xmm4,%0" : "=m" (t[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (s[i]));
+		asm volatile ("movntdq %%xmm4,%0" : "=m" (t[i]));
 	}
 
 	raid_asm_clobber_xmm16();
@@ -1448,96 +1448,96 @@ void raid_gen5_avx2ext(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("vpxor %ymm8,%ymm8,%ymm8");
-	asm volatile("vbroadcasti128 %0,%%ymm14" : : "m" (gfconst16.poly[0]));
-	asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
+	asm volatile ("vpxor %ymm8,%ymm8,%ymm8");
+	asm volatile ("vbroadcasti128 %0,%%ymm14" : : "m" (gfconst16.poly[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 32) {
 		/* last disk without the by two multiplication */
-		asm volatile("vmovdqa %0,%%ymm10" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm10" : : "m" (v[l][i]));
 
-		asm volatile("vmovdqa %ymm10,%ymm0");
-		asm volatile("vmovdqa %ymm10,%ymm1");
+		asm volatile ("vmovdqa %ymm10,%ymm0");
+		asm volatile ("vmovdqa %ymm10,%ymm1");
 
-		asm volatile("vpsrlw  $4,%ymm10,%ymm11");
-		asm volatile("vpand   %ymm15,%ymm10,%ymm10");
-		asm volatile("vpand   %ymm15,%ymm11,%ymm11");
+		asm volatile ("vpsrlw  $4,%ymm10,%ymm11");
+		asm volatile ("vpand   %ymm15,%ymm10,%ymm10");
+		asm volatile ("vpand   %ymm15,%ymm11,%ymm11");
 
-		asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm2,%ymm2");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm2,%ymm2");
+		asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm2,%ymm2");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm2,%ymm2");
 
-		asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm3,%ymm3");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm3,%ymm3");
+		asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm3,%ymm3");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm3,%ymm3");
 
-		asm volatile("vbroadcasti128 %0,%%ymm4" : : "m" (gfgenpshufb[l][2][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][2][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm4,%ymm4");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm4,%ymm4");
+		asm volatile ("vbroadcasti128 %0,%%ymm4" : : "m" (gfgenpshufb[l][2][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][2][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm4,%ymm4");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm4,%ymm4");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("vmovdqa %0,%%ymm10" : : "m" (v[d][i]));
+			asm volatile ("vmovdqa %0,%%ymm10" : : "m" (v[d][i]));
 
-			asm volatile("vpcmpgtb %ymm1,%ymm8,%ymm11");
-			asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-			asm volatile("vpand %ymm14,%ymm11,%ymm11");
-			asm volatile("vpxor %ymm11,%ymm1,%ymm1");
+			asm volatile ("vpcmpgtb %ymm1,%ymm8,%ymm11");
+			asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+			asm volatile ("vpand %ymm14,%ymm11,%ymm11");
+			asm volatile ("vpxor %ymm11,%ymm1,%ymm1");
 
-			asm volatile("vpxor %ymm10,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm10,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm10,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm10,%ymm1,%ymm1");
 
-			asm volatile("vpsrlw  $4,%ymm10,%ymm11");
-			asm volatile("vpand   %ymm15,%ymm10,%ymm10");
-			asm volatile("vpand   %ymm15,%ymm11,%ymm11");
+			asm volatile ("vpsrlw  $4,%ymm10,%ymm11");
+			asm volatile ("vpand   %ymm15,%ymm10,%ymm10");
+			asm volatile ("vpand   %ymm15,%ymm11,%ymm11");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm2,%ymm2");
-			asm volatile("vpxor   %ymm13,%ymm2,%ymm2");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm2,%ymm2");
+			asm volatile ("vpxor   %ymm13,%ymm2,%ymm2");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm3,%ymm3");
-			asm volatile("vpxor   %ymm13,%ymm3,%ymm3");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm3,%ymm3");
+			asm volatile ("vpxor   %ymm13,%ymm3,%ymm3");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][2][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][2][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm4,%ymm4");
-			asm volatile("vpxor   %ymm13,%ymm4,%ymm4");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][2][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][2][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm4,%ymm4");
+			asm volatile ("vpxor   %ymm13,%ymm4,%ymm4");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("vmovdqa %0,%%ymm10" : : "m" (v[0][i]));
+		asm volatile ("vmovdqa %0,%%ymm10" : : "m" (v[0][i]));
 
-		asm volatile("vpcmpgtb %ymm1,%ymm8,%ymm11");
-		asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-		asm volatile("vpand %ymm14,%ymm11,%ymm11");
-		asm volatile("vpxor %ymm11,%ymm1,%ymm1");
+		asm volatile ("vpcmpgtb %ymm1,%ymm8,%ymm11");
+		asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+		asm volatile ("vpand %ymm14,%ymm11,%ymm11");
+		asm volatile ("vpxor %ymm11,%ymm1,%ymm1");
 
-		asm volatile("vpxor %ymm10,%ymm0,%ymm0");
-		asm volatile("vpxor %ymm10,%ymm1,%ymm1");
-		asm volatile("vpxor %ymm10,%ymm2,%ymm2");
-		asm volatile("vpxor %ymm10,%ymm3,%ymm3");
-		asm volatile("vpxor %ymm10,%ymm4,%ymm4");
+		asm volatile ("vpxor %ymm10,%ymm0,%ymm0");
+		asm volatile ("vpxor %ymm10,%ymm1,%ymm1");
+		asm volatile ("vpxor %ymm10,%ymm2,%ymm2");
+		asm volatile ("vpxor %ymm10,%ymm3,%ymm3");
+		asm volatile ("vpxor %ymm10,%ymm4,%ymm4");
 
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (q[i]));
-		asm volatile("vmovntdq %%ymm2,%0" : "=m" (r[i]));
-		asm volatile("vmovntdq %%ymm3,%0" : "=m" (s[i]));
-		asm volatile("vmovntdq %%ymm4,%0" : "=m" (t[i]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (q[i]));
+		asm volatile ("vmovntdq %%ymm2,%0" : "=m" (r[i]));
+		asm volatile ("vmovntdq %%ymm3,%0" : "=m" (s[i]));
+		asm volatile ("vmovntdq %%ymm4,%0" : "=m" (t[i]));
 	}
 
 	raid_asm_clobber_ymm16();
@@ -1588,120 +1588,120 @@ void raid_gen6_ssse3(int nd, size_t size, void **vv)
 	/* generic case with at least two data disks */
 	for (i = 0; i < size; i += 16) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[l][i]));
 
-		asm volatile("movdqa %%xmm4,%0" : "=m" (p0[0]));
-		asm volatile("movdqa %%xmm4,%0" : "=m" (q0[0]));
+		asm volatile ("movdqa %%xmm4,%0" : "=m" (p0[0]));
+		asm volatile ("movdqa %%xmm4,%0" : "=m" (q0[0]));
 
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
-		asm volatile("movdqa %xmm4,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+		asm volatile ("movdqa %xmm4,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
 
-		asm volatile("movdqa %0,%%xmm0" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("pshufb %xmm4,%xmm0");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm0");
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm0");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm0");
 
-		asm volatile("movdqa %0,%%xmm1" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("pshufb %xmm4,%xmm1");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm1");
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm1");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm1");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][2][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][2][1][0]));
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm2");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][2][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][2][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm2");
 
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][3][0][0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][3][1][0]));
-		asm volatile("pshufb %xmm4,%xmm3");
-		asm volatile("pshufb %xmm5,%xmm7");
-		asm volatile("pxor   %xmm7,%xmm3");
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][3][0][0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[l][3][1][0]));
+		asm volatile ("pshufb %xmm4,%xmm3");
+		asm volatile ("pshufb %xmm5,%xmm7");
+		asm volatile ("pxor   %xmm7,%xmm3");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm5" : : "m" (p0[0]));
-			asm volatile("movdqa %0,%%xmm6" : : "m" (q0[0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+			asm volatile ("movdqa %0,%%xmm5" : : "m" (p0[0]));
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (q0[0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
 
-			asm volatile("pxor %xmm4,%xmm4");
-			asm volatile("pcmpgtb %xmm6,%xmm4");
-			asm volatile("paddb %xmm6,%xmm6");
-			asm volatile("pand %xmm7,%xmm4");
-			asm volatile("pxor %xmm4,%xmm6");
+			asm volatile ("pxor %xmm4,%xmm4");
+			asm volatile ("pcmpgtb %xmm6,%xmm4");
+			asm volatile ("paddb %xmm6,%xmm6");
+			asm volatile ("pand %xmm7,%xmm4");
+			asm volatile ("pxor %xmm4,%xmm6");
 
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
 
-			asm volatile("pxor %xmm4,%xmm5");
-			asm volatile("pxor %xmm4,%xmm6");
-			asm volatile("movdqa %%xmm5,%0" : "=m" (p0[0]));
-			asm volatile("movdqa %%xmm6,%0" : "=m" (q0[0]));
+			asm volatile ("pxor %xmm4,%xmm5");
+			asm volatile ("pxor %xmm4,%xmm6");
+			asm volatile ("movdqa %%xmm5,%0" : "=m" (p0[0]));
+			asm volatile ("movdqa %%xmm6,%0" : "=m" (q0[0]));
 
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
-			asm volatile("movdqa %xmm4,%xmm5");
-			asm volatile("psrlw  $4,%xmm5");
-			asm volatile("pand   %xmm7,%xmm4");
-			asm volatile("pand   %xmm7,%xmm5");
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+			asm volatile ("movdqa %xmm4,%xmm5");
+			asm volatile ("psrlw  $4,%xmm5");
+			asm volatile ("pand   %xmm7,%xmm4");
+			asm volatile ("pand   %xmm7,%xmm5");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm0");
-			asm volatile("pxor   %xmm7,%xmm0");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm0");
+			asm volatile ("pxor   %xmm7,%xmm0");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm1");
-			asm volatile("pxor   %xmm7,%xmm1");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm1");
+			asm volatile ("pxor   %xmm7,%xmm1");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][2][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][2][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm2");
-			asm volatile("pxor   %xmm7,%xmm2");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][2][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][2][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm2");
+			asm volatile ("pxor   %xmm7,%xmm2");
 
-			asm volatile("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][3][0][0]));
-			asm volatile("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][3][1][0]));
-			asm volatile("pshufb %xmm4,%xmm6");
-			asm volatile("pshufb %xmm5,%xmm7");
-			asm volatile("pxor   %xmm6,%xmm3");
-			asm volatile("pxor   %xmm7,%xmm3");
+			asm volatile ("movdqa %0,%%xmm6" : : "m" (gfgenpshufb[d][3][0][0]));
+			asm volatile ("movdqa %0,%%xmm7" : : "m" (gfgenpshufb[d][3][1][0]));
+			asm volatile ("pshufb %xmm4,%xmm6");
+			asm volatile ("pshufb %xmm5,%xmm7");
+			asm volatile ("pxor   %xmm6,%xmm3");
+			asm volatile ("pxor   %xmm7,%xmm3");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm5" : : "m" (p0[0]));
-		asm volatile("movdqa %0,%%xmm6" : : "m" (q0[0]));
-		asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
+		asm volatile ("movdqa %0,%%xmm5" : : "m" (p0[0]));
+		asm volatile ("movdqa %0,%%xmm6" : : "m" (q0[0]));
+		asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.poly[0]));
 
-		asm volatile("pxor %xmm4,%xmm4");
-		asm volatile("pcmpgtb %xmm6,%xmm4");
-		asm volatile("paddb %xmm6,%xmm6");
-		asm volatile("pand %xmm7,%xmm4");
-		asm volatile("pxor %xmm4,%xmm6");
+		asm volatile ("pxor %xmm4,%xmm4");
+		asm volatile ("pcmpgtb %xmm6,%xmm4");
+		asm volatile ("paddb %xmm6,%xmm6");
+		asm volatile ("pand %xmm7,%xmm4");
+		asm volatile ("pxor %xmm4,%xmm6");
 
-		asm volatile("movdqa %0,%%xmm4" : : "m" (v[0][i]));
-		asm volatile("pxor %xmm4,%xmm0");
-		asm volatile("pxor %xmm4,%xmm1");
-		asm volatile("pxor %xmm4,%xmm2");
-		asm volatile("pxor %xmm4,%xmm3");
-		asm volatile("pxor %xmm4,%xmm5");
-		asm volatile("pxor %xmm4,%xmm6");
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (v[0][i]));
+		asm volatile ("pxor %xmm4,%xmm0");
+		asm volatile ("pxor %xmm4,%xmm1");
+		asm volatile ("pxor %xmm4,%xmm2");
+		asm volatile ("pxor %xmm4,%xmm3");
+		asm volatile ("pxor %xmm4,%xmm5");
+		asm volatile ("pxor %xmm4,%xmm6");
 
-		asm volatile("movntdq %%xmm5,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm6,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm0,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (s[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (t[i]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (u[i]));
+		asm volatile ("movntdq %%xmm5,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm6,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (s[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (t[i]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (u[i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -1746,114 +1746,114 @@ void raid_gen6_ssse3ext(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("movdqa %0,%%xmm14" : : "m" (gfconst16.poly[0]));
-	asm volatile("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm14" : : "m" (gfconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm15" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 16) {
 		/* last disk without the by two multiplication */
-		asm volatile("movdqa %0,%%xmm10" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm10" : : "m" (v[l][i]));
 
-		asm volatile("movdqa %xmm10,%xmm0");
-		asm volatile("movdqa %xmm10,%xmm1");
+		asm volatile ("movdqa %xmm10,%xmm0");
+		asm volatile ("movdqa %xmm10,%xmm1");
 
-		asm volatile("movdqa %xmm10,%xmm11");
-		asm volatile("psrlw  $4,%xmm11");
-		asm volatile("pand   %xmm15,%xmm10");
-		asm volatile("pand   %xmm15,%xmm11");
+		asm volatile ("movdqa %xmm10,%xmm11");
+		asm volatile ("psrlw  $4,%xmm11");
+		asm volatile ("pand   %xmm15,%xmm10");
+		asm volatile ("pand   %xmm15,%xmm11");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("pshufb %xmm10,%xmm2");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm2");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm2");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm2");
 
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("pshufb %xmm10,%xmm3");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm3");
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm3");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm3");
 
-		asm volatile("movdqa %0,%%xmm4" : : "m" (gfgenpshufb[l][2][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][2][1][0]));
-		asm volatile("pshufb %xmm10,%xmm4");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm4");
+		asm volatile ("movdqa %0,%%xmm4" : : "m" (gfgenpshufb[l][2][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][2][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm4");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm4");
 
-		asm volatile("movdqa %0,%%xmm5" : : "m" (gfgenpshufb[l][3][0][0]));
-		asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][3][1][0]));
-		asm volatile("pshufb %xmm10,%xmm5");
-		asm volatile("pshufb %xmm11,%xmm13");
-		asm volatile("pxor   %xmm13,%xmm5");
+		asm volatile ("movdqa %0,%%xmm5" : : "m" (gfgenpshufb[l][3][0][0]));
+		asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[l][3][1][0]));
+		asm volatile ("pshufb %xmm10,%xmm5");
+		asm volatile ("pshufb %xmm11,%xmm13");
+		asm volatile ("pxor   %xmm13,%xmm5");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("movdqa %0,%%xmm10" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm10" : : "m" (v[d][i]));
 
-			asm volatile("pxor %xmm11,%xmm11");
-			asm volatile("pcmpgtb %xmm1,%xmm11");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("pand %xmm14,%xmm11");
-			asm volatile("pxor %xmm11,%xmm1");
+			asm volatile ("pxor %xmm11,%xmm11");
+			asm volatile ("pcmpgtb %xmm1,%xmm11");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("pand %xmm14,%xmm11");
+			asm volatile ("pxor %xmm11,%xmm1");
 
-			asm volatile("pxor %xmm10,%xmm0");
-			asm volatile("pxor %xmm10,%xmm1");
+			asm volatile ("pxor %xmm10,%xmm0");
+			asm volatile ("pxor %xmm10,%xmm1");
 
-			asm volatile("movdqa %xmm10,%xmm11");
-			asm volatile("psrlw  $4,%xmm11");
-			asm volatile("pand   %xmm15,%xmm10");
-			asm volatile("pand   %xmm15,%xmm11");
+			asm volatile ("movdqa %xmm10,%xmm11");
+			asm volatile ("psrlw  $4,%xmm11");
+			asm volatile ("pand   %xmm15,%xmm10");
+			asm volatile ("pand   %xmm15,%xmm11");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm2");
-			asm volatile("pxor   %xmm13,%xmm2");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm2");
+			asm volatile ("pxor   %xmm13,%xmm2");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm3");
-			asm volatile("pxor   %xmm13,%xmm3");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm3");
+			asm volatile ("pxor   %xmm13,%xmm3");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][2][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][2][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm4");
-			asm volatile("pxor   %xmm13,%xmm4");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][2][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][2][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm4");
+			asm volatile ("pxor   %xmm13,%xmm4");
 
-			asm volatile("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][3][0][0]));
-			asm volatile("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][3][1][0]));
-			asm volatile("pshufb %xmm10,%xmm12");
-			asm volatile("pshufb %xmm11,%xmm13");
-			asm volatile("pxor   %xmm12,%xmm5");
-			asm volatile("pxor   %xmm13,%xmm5");
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (gfgenpshufb[d][3][0][0]));
+			asm volatile ("movdqa %0,%%xmm13" : : "m" (gfgenpshufb[d][3][1][0]));
+			asm volatile ("pshufb %xmm10,%xmm12");
+			asm volatile ("pshufb %xmm11,%xmm13");
+			asm volatile ("pxor   %xmm12,%xmm5");
+			asm volatile ("pxor   %xmm13,%xmm5");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("movdqa %0,%%xmm10" : : "m" (v[0][i]));
+		asm volatile ("movdqa %0,%%xmm10" : : "m" (v[0][i]));
 
-		asm volatile("pxor %xmm11,%xmm11");
-		asm volatile("pcmpgtb %xmm1,%xmm11");
-		asm volatile("paddb %xmm1,%xmm1");
-		asm volatile("pand %xmm14,%xmm11");
-		asm volatile("pxor %xmm11,%xmm1");
+		asm volatile ("pxor %xmm11,%xmm11");
+		asm volatile ("pcmpgtb %xmm1,%xmm11");
+		asm volatile ("paddb %xmm1,%xmm1");
+		asm volatile ("pand %xmm14,%xmm11");
+		asm volatile ("pxor %xmm11,%xmm1");
 
-		asm volatile("pxor %xmm10,%xmm0");
-		asm volatile("pxor %xmm10,%xmm1");
-		asm volatile("pxor %xmm10,%xmm2");
-		asm volatile("pxor %xmm10,%xmm3");
-		asm volatile("pxor %xmm10,%xmm4");
-		asm volatile("pxor %xmm10,%xmm5");
+		asm volatile ("pxor %xmm10,%xmm0");
+		asm volatile ("pxor %xmm10,%xmm1");
+		asm volatile ("pxor %xmm10,%xmm2");
+		asm volatile ("pxor %xmm10,%xmm3");
+		asm volatile ("pxor %xmm10,%xmm4");
+		asm volatile ("pxor %xmm10,%xmm5");
 
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm3,%0" : "=m" (s[i]));
-		asm volatile("movntdq %%xmm4,%0" : "=m" (t[i]));
-		asm volatile("movntdq %%xmm5,%0" : "=m" (u[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm3,%0" : "=m" (s[i]));
+		asm volatile ("movntdq %%xmm4,%0" : "=m" (t[i]));
+		asm volatile ("movntdq %%xmm5,%0" : "=m" (u[i]));
 	}
 
 	raid_asm_clobber_xmm16();
@@ -1898,111 +1898,111 @@ void raid_gen6_avx2ext(int nd, size_t size, void **vv)
 	raid_asm_begin();
 
 	/* generic case with at least two data disks */
-	asm volatile("vpxor %ymm8,%ymm8,%ymm8");
-	asm volatile("vbroadcasti128 %0,%%ymm14" : : "m" (gfconst16.poly[0]));
-	asm volatile("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
+	asm volatile ("vpxor %ymm8,%ymm8,%ymm8");
+	asm volatile ("vbroadcasti128 %0,%%ymm14" : : "m" (gfconst16.poly[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm15" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 32) {
 		/* last disk without the by two multiplication */
-		asm volatile("vmovdqa %0,%%ymm10" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm10" : : "m" (v[l][i]));
 
-		asm volatile("vmovdqa %ymm10,%ymm0");
-		asm volatile("vmovdqa %ymm10,%ymm1");
+		asm volatile ("vmovdqa %ymm10,%ymm0");
+		asm volatile ("vmovdqa %ymm10,%ymm1");
 
-		asm volatile("vpsrlw  $4,%ymm10,%ymm11");
-		asm volatile("vpand   %ymm15,%ymm10,%ymm10");
-		asm volatile("vpand   %ymm15,%ymm11,%ymm11");
+		asm volatile ("vpsrlw  $4,%ymm10,%ymm11");
+		asm volatile ("vpand   %ymm15,%ymm10,%ymm10");
+		asm volatile ("vpand   %ymm15,%ymm11,%ymm11");
 
-		asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfgenpshufb[l][0][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][0][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm2,%ymm2");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm2,%ymm2");
+		asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfgenpshufb[l][0][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][0][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm2,%ymm2");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm2,%ymm2");
 
-		asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfgenpshufb[l][1][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][1][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm3,%ymm3");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm3,%ymm3");
+		asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfgenpshufb[l][1][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][1][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm3,%ymm3");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm3,%ymm3");
 
-		asm volatile("vbroadcasti128 %0,%%ymm4" : : "m" (gfgenpshufb[l][2][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][2][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm4,%ymm4");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm4,%ymm4");
+		asm volatile ("vbroadcasti128 %0,%%ymm4" : : "m" (gfgenpshufb[l][2][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][2][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm4,%ymm4");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm4,%ymm4");
 
-		asm volatile("vbroadcasti128 %0,%%ymm5" : : "m" (gfgenpshufb[l][3][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][3][1][0]));
-		asm volatile("vpshufb %ymm10,%ymm5,%ymm5");
-		asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-		asm volatile("vpxor   %ymm13,%ymm5,%ymm5");
+		asm volatile ("vbroadcasti128 %0,%%ymm5" : : "m" (gfgenpshufb[l][3][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[l][3][1][0]));
+		asm volatile ("vpshufb %ymm10,%ymm5,%ymm5");
+		asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+		asm volatile ("vpxor   %ymm13,%ymm5,%ymm5");
 
 		/* intermediate disks */
 		for (d = l-1; d > 0; --d) {
-			asm volatile("vmovdqa %0,%%ymm10" : : "m" (v[d][i]));
+			asm volatile ("vmovdqa %0,%%ymm10" : : "m" (v[d][i]));
 
-			asm volatile("vpcmpgtb %ymm1,%ymm8,%ymm11");
-			asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-			asm volatile("vpand %ymm14,%ymm11,%ymm11");
-			asm volatile("vpxor %ymm11,%ymm1,%ymm1");
+			asm volatile ("vpcmpgtb %ymm1,%ymm8,%ymm11");
+			asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+			asm volatile ("vpand %ymm14,%ymm11,%ymm11");
+			asm volatile ("vpxor %ymm11,%ymm1,%ymm1");
 
-			asm volatile("vpxor %ymm10,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm10,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm10,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm10,%ymm1,%ymm1");
 
-			asm volatile("vpsrlw  $4,%ymm10,%ymm11");
-			asm volatile("vpand   %ymm15,%ymm10,%ymm10");
-			asm volatile("vpand   %ymm15,%ymm11,%ymm11");
+			asm volatile ("vpsrlw  $4,%ymm10,%ymm11");
+			asm volatile ("vpand   %ymm15,%ymm10,%ymm10");
+			asm volatile ("vpand   %ymm15,%ymm11,%ymm11");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][0][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][0][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm2,%ymm2");
-			asm volatile("vpxor   %ymm13,%ymm2,%ymm2");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][0][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][0][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm2,%ymm2");
+			asm volatile ("vpxor   %ymm13,%ymm2,%ymm2");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][1][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][1][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm3,%ymm3");
-			asm volatile("vpxor   %ymm13,%ymm3,%ymm3");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][1][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][1][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm3,%ymm3");
+			asm volatile ("vpxor   %ymm13,%ymm3,%ymm3");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][2][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][2][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm4,%ymm4");
-			asm volatile("vpxor   %ymm13,%ymm4,%ymm4");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][2][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][2][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm4,%ymm4");
+			asm volatile ("vpxor   %ymm13,%ymm4,%ymm4");
 
-			asm volatile("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][3][0][0]));
-			asm volatile("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][3][1][0]));
-			asm volatile("vpshufb %ymm10,%ymm12,%ymm12");
-			asm volatile("vpshufb %ymm11,%ymm13,%ymm13");
-			asm volatile("vpxor   %ymm12,%ymm5,%ymm5");
-			asm volatile("vpxor   %ymm13,%ymm5,%ymm5");
+			asm volatile ("vbroadcasti128 %0,%%ymm12" : : "m" (gfgenpshufb[d][3][0][0]));
+			asm volatile ("vbroadcasti128 %0,%%ymm13" : : "m" (gfgenpshufb[d][3][1][0]));
+			asm volatile ("vpshufb %ymm10,%ymm12,%ymm12");
+			asm volatile ("vpshufb %ymm11,%ymm13,%ymm13");
+			asm volatile ("vpxor   %ymm12,%ymm5,%ymm5");
+			asm volatile ("vpxor   %ymm13,%ymm5,%ymm5");
 		}
 
 		/* first disk with all coefficients at 1 */
-		asm volatile("vmovdqa %0,%%ymm10" : : "m" (v[0][i]));
+		asm volatile ("vmovdqa %0,%%ymm10" : : "m" (v[0][i]));
 
-		asm volatile("vpcmpgtb %ymm1,%ymm8,%ymm11");
-		asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-		asm volatile("vpand %ymm14,%ymm11,%ymm11");
-		asm volatile("vpxor %ymm11,%ymm1,%ymm1");
+		asm volatile ("vpcmpgtb %ymm1,%ymm8,%ymm11");
+		asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+		asm volatile ("vpand %ymm14,%ymm11,%ymm11");
+		asm volatile ("vpxor %ymm11,%ymm1,%ymm1");
 
-		asm volatile("vpxor %ymm10,%ymm0,%ymm0");
-		asm volatile("vpxor %ymm10,%ymm1,%ymm1");
-		asm volatile("vpxor %ymm10,%ymm2,%ymm2");
-		asm volatile("vpxor %ymm10,%ymm3,%ymm3");
-		asm volatile("vpxor %ymm10,%ymm4,%ymm4");
-		asm volatile("vpxor %ymm10,%ymm5,%ymm5");
+		asm volatile ("vpxor %ymm10,%ymm0,%ymm0");
+		asm volatile ("vpxor %ymm10,%ymm1,%ymm1");
+		asm volatile ("vpxor %ymm10,%ymm2,%ymm2");
+		asm volatile ("vpxor %ymm10,%ymm3,%ymm3");
+		asm volatile ("vpxor %ymm10,%ymm4,%ymm4");
+		asm volatile ("vpxor %ymm10,%ymm5,%ymm5");
 
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (q[i]));
-		asm volatile("vmovntdq %%ymm2,%0" : "=m" (r[i]));
-		asm volatile("vmovntdq %%ymm3,%0" : "=m" (s[i]));
-		asm volatile("vmovntdq %%ymm4,%0" : "=m" (t[i]));
-		asm volatile("vmovntdq %%ymm5,%0" : "=m" (u[i]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (q[i]));
+		asm volatile ("vmovntdq %%ymm2,%0" : "=m" (r[i]));
+		asm volatile ("vmovntdq %%ymm3,%0" : "=m" (s[i]));
+		asm volatile ("vmovntdq %%ymm4,%0" : "=m" (t[i]));
+		asm volatile ("vmovntdq %%ymm5,%0" : "=m" (u[i]));
 	}
 
 	raid_asm_clobber_ymm16();
@@ -2046,24 +2046,24 @@ void raid_rec1_ssse3(int nr, int *id, int *ip, int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
-	asm volatile("movdqa %0,%%xmm4" : : "m" (gfmulpshufb[V][0][0]));
-	asm volatile("movdqa %0,%%xmm5" : : "m" (gfmulpshufb[V][1][0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm4" : : "m" (gfmulpshufb[V][0][0]));
+	asm volatile ("movdqa %0,%%xmm5" : : "m" (gfmulpshufb[V][1][0]));
 
 	for (i = 0; i < size; i += 16) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (p[i]));
-		asm volatile("movdqa %0,%%xmm1" : : "m" (pa[i]));
-		asm volatile("movdqa %xmm4,%xmm2");
-		asm volatile("movdqa %xmm5,%xmm3");
-		asm volatile("pxor   %xmm0,%xmm1");
-		asm volatile("movdqa %xmm1,%xmm0");
-		asm volatile("psrlw  $4,%xmm1");
-		asm volatile("pand   %xmm7,%xmm0");
-		asm volatile("pand   %xmm7,%xmm1");
-		asm volatile("pshufb %xmm0,%xmm2");
-		asm volatile("pshufb %xmm1,%xmm3");
-		asm volatile("pxor   %xmm3,%xmm2");
-		asm volatile("movdqa %%xmm2,%0" : "=m" (pa[i]));
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (p[i]));
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (pa[i]));
+		asm volatile ("movdqa %xmm4,%xmm2");
+		asm volatile ("movdqa %xmm5,%xmm3");
+		asm volatile ("pxor   %xmm0,%xmm1");
+		asm volatile ("movdqa %xmm1,%xmm0");
+		asm volatile ("psrlw  $4,%xmm1");
+		asm volatile ("pand   %xmm7,%xmm0");
+		asm volatile ("pand   %xmm7,%xmm1");
+		asm volatile ("pshufb %xmm0,%xmm2");
+		asm volatile ("pshufb %xmm1,%xmm3");
+		asm volatile ("pxor   %xmm3,%xmm2");
+		asm volatile ("movdqa %%xmm2,%0" : "=m" (pa[i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -2107,71 +2107,71 @@ void raid_rec2_ssse3(int nr, int *id, int *ip, int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 16) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (p[0][i]));
-		asm volatile("movdqa %0,%%xmm2" : : "m" (pa[0][i]));
-		asm volatile("movdqa %0,%%xmm1" : : "m" (p[1][i]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (pa[1][i]));
-		asm volatile("pxor   %xmm2,%xmm0");
-		asm volatile("pxor   %xmm3,%xmm1");
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (p[0][i]));
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (pa[0][i]));
+		asm volatile ("movdqa %0,%%xmm1" : : "m" (p[1][i]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (pa[1][i]));
+		asm volatile ("pxor   %xmm2,%xmm0");
+		asm volatile ("pxor   %xmm3,%xmm1");
 
-		asm volatile("pxor %xmm6,%xmm6");
+		asm volatile ("pxor %xmm6,%xmm6");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[0]][0][0]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[0]][1][0]));
-		asm volatile("movdqa %xmm0,%xmm4");
-		asm volatile("movdqa %xmm0,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm3");
-		asm volatile("pxor   %xmm2,%xmm6");
-		asm volatile("pxor   %xmm3,%xmm6");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[0]][0][0]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[0]][1][0]));
+		asm volatile ("movdqa %xmm0,%xmm4");
+		asm volatile ("movdqa %xmm0,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm3");
+		asm volatile ("pxor   %xmm2,%xmm6");
+		asm volatile ("pxor   %xmm3,%xmm6");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[1]][0][0]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[1]][1][0]));
-		asm volatile("movdqa %xmm1,%xmm4");
-		asm volatile("movdqa %xmm1,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm3");
-		asm volatile("pxor   %xmm2,%xmm6");
-		asm volatile("pxor   %xmm3,%xmm6");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[1]][0][0]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[1]][1][0]));
+		asm volatile ("movdqa %xmm1,%xmm4");
+		asm volatile ("movdqa %xmm1,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm3");
+		asm volatile ("pxor   %xmm2,%xmm6");
+		asm volatile ("pxor   %xmm3,%xmm6");
 
-		asm volatile("movdqa %%xmm6,%0" : "=m" (pa[0][i]));
+		asm volatile ("movdqa %%xmm6,%0" : "=m" (pa[0][i]));
 
-		asm volatile("pxor %xmm6,%xmm6");
+		asm volatile ("pxor %xmm6,%xmm6");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[2]][0][0]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[2]][1][0]));
-		asm volatile("movdqa %xmm0,%xmm4");
-		asm volatile("movdqa %xmm0,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm3");
-		asm volatile("pxor   %xmm2,%xmm6");
-		asm volatile("pxor   %xmm3,%xmm6");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[2]][0][0]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[2]][1][0]));
+		asm volatile ("movdqa %xmm0,%xmm4");
+		asm volatile ("movdqa %xmm0,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm3");
+		asm volatile ("pxor   %xmm2,%xmm6");
+		asm volatile ("pxor   %xmm3,%xmm6");
 
-		asm volatile("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[3]][0][0]));
-		asm volatile("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[3]][1][0]));
-		asm volatile("movdqa %xmm1,%xmm4");
-		asm volatile("movdqa %xmm1,%xmm5");
-		asm volatile("psrlw  $4,%xmm5");
-		asm volatile("pand   %xmm7,%xmm4");
-		asm volatile("pand   %xmm7,%xmm5");
-		asm volatile("pshufb %xmm4,%xmm2");
-		asm volatile("pshufb %xmm5,%xmm3");
-		asm volatile("pxor   %xmm2,%xmm6");
-		asm volatile("pxor   %xmm3,%xmm6");
+		asm volatile ("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[V[3]][0][0]));
+		asm volatile ("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[V[3]][1][0]));
+		asm volatile ("movdqa %xmm1,%xmm4");
+		asm volatile ("movdqa %xmm1,%xmm5");
+		asm volatile ("psrlw  $4,%xmm5");
+		asm volatile ("pand   %xmm7,%xmm4");
+		asm volatile ("pand   %xmm7,%xmm5");
+		asm volatile ("pshufb %xmm4,%xmm2");
+		asm volatile ("pshufb %xmm5,%xmm3");
+		asm volatile ("pxor   %xmm2,%xmm6");
+		asm volatile ("pxor   %xmm3,%xmm6");
 
-		asm volatile("movdqa %%xmm6,%0" : "=m" (pa[1][i]));
+		asm volatile ("movdqa %%xmm6,%0" : "=m" (pa[1][i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -2213,42 +2213,42 @@ void raid_recX_ssse3(int nr, int *id, int *ip, int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 16) {
 		uint8_t PD[RAID_PARITY_MAX][16] __aligned(16);
 
 		/* delta */
 		for (j = 0; j < N; ++j) {
-			asm volatile("movdqa %0,%%xmm0" : : "m" (p[j][i]));
-			asm volatile("movdqa %0,%%xmm1" : : "m" (pa[j][i]));
-			asm volatile("pxor   %xmm1,%xmm0");
-			asm volatile("movdqa %%xmm0,%0" : "=m" (PD[j][0]));
+			asm volatile ("movdqa %0,%%xmm0" : : "m" (p[j][i]));
+			asm volatile ("movdqa %0,%%xmm1" : : "m" (pa[j][i]));
+			asm volatile ("pxor   %xmm1,%xmm0");
+			asm volatile ("movdqa %%xmm0,%0" : "=m" (PD[j][0]));
 		}
 
 		/* reconstruct */
 		for (j = 0; j < N; ++j) {
-			asm volatile("pxor %xmm0,%xmm0");
-			asm volatile("pxor %xmm1,%xmm1");
+			asm volatile ("pxor %xmm0,%xmm0");
+			asm volatile ("pxor %xmm1,%xmm1");
 
 			for (k = 0; k < N; ++k) {
 				uint8_t m = V[j*N+k];
 
-				asm volatile("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[m][0][0]));
-				asm volatile("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[m][1][0]));
-				asm volatile("movdqa %0,%%xmm4" : : "m" (PD[k][0]));
-				asm volatile("movdqa %xmm4,%xmm5");
-				asm volatile("psrlw  $4,%xmm5");
-				asm volatile("pand   %xmm7,%xmm4");
-				asm volatile("pand   %xmm7,%xmm5");
-				asm volatile("pshufb %xmm4,%xmm2");
-				asm volatile("pshufb %xmm5,%xmm3");
-				asm volatile("pxor   %xmm2,%xmm0");
-				asm volatile("pxor   %xmm3,%xmm1");
+				asm volatile ("movdqa %0,%%xmm2" : : "m" (gfmulpshufb[m][0][0]));
+				asm volatile ("movdqa %0,%%xmm3" : : "m" (gfmulpshufb[m][1][0]));
+				asm volatile ("movdqa %0,%%xmm4" : : "m" (PD[k][0]));
+				asm volatile ("movdqa %xmm4,%xmm5");
+				asm volatile ("psrlw  $4,%xmm5");
+				asm volatile ("pand   %xmm7,%xmm4");
+				asm volatile ("pand   %xmm7,%xmm5");
+				asm volatile ("pshufb %xmm4,%xmm2");
+				asm volatile ("pshufb %xmm5,%xmm3");
+				asm volatile ("pxor   %xmm2,%xmm0");
+				asm volatile ("pxor   %xmm3,%xmm1");
 			}
 
-			asm volatile("pxor %xmm1,%xmm0");
-			asm volatile("movdqa %%xmm0,%0" : "=m" (pa[j][i]));
+			asm volatile ("pxor %xmm1,%xmm0");
+			asm volatile ("movdqa %%xmm0,%0" : "=m" (pa[j][i]));
 		}
 	}
 
@@ -2294,21 +2294,21 @@ void raid_rec1_avx2(int nr, int *id, int *ip, int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.low4[0]));
-	asm volatile("vbroadcasti128 %0,%%ymm4" : : "m" (gfmulpshufb[V][0][0]));
-	asm volatile("vbroadcasti128 %0,%%ymm5" : : "m" (gfmulpshufb[V][1][0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm4" : : "m" (gfmulpshufb[V][0][0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm5" : : "m" (gfmulpshufb[V][1][0]));
 
 	for (i = 0; i < size; i += 32) {
-		asm volatile("vmovdqa %0,%%ymm0" : : "m" (p[i]));
-		asm volatile("vmovdqa %0,%%ymm1" : : "m" (pa[i]));
-		asm volatile("vpxor   %ymm1,%ymm0,%ymm0");
-		asm volatile("vpsrlw  $4,%ymm0,%ymm1");
-		asm volatile("vpand   %ymm7,%ymm0,%ymm0");
-		asm volatile("vpand   %ymm7,%ymm1,%ymm1");
-		asm volatile("vpshufb %ymm0,%ymm4,%ymm2");
-		asm volatile("vpshufb %ymm1,%ymm5,%ymm3");
-		asm volatile("vpxor   %ymm3,%ymm2,%ymm2");
-		asm volatile("vmovdqa %%ymm2,%0" : "=m" (pa[i]));
+		asm volatile ("vmovdqa %0,%%ymm0" : : "m" (p[i]));
+		asm volatile ("vmovdqa %0,%%ymm1" : : "m" (pa[i]));
+		asm volatile ("vpxor   %ymm1,%ymm0,%ymm0");
+		asm volatile ("vpsrlw  $4,%ymm0,%ymm1");
+		asm volatile ("vpand   %ymm7,%ymm0,%ymm0");
+		asm volatile ("vpand   %ymm7,%ymm1,%ymm1");
+		asm volatile ("vpshufb %ymm0,%ymm4,%ymm2");
+		asm volatile ("vpshufb %ymm1,%ymm5,%ymm3");
+		asm volatile ("vpxor   %ymm3,%ymm2,%ymm2");
+		asm volatile ("vmovdqa %%ymm2,%0" : "=m" (pa[i]));
 	}
 
 	raid_asm_clobber_ymm8();
@@ -2352,63 +2352,63 @@ void raid_rec2_avx2(int nr, int *id, int *ip, int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 32) {
-		asm volatile("vmovdqa %0,%%ymm0" : : "m" (p[0][i]));
-		asm volatile("vmovdqa %0,%%ymm2" : : "m" (pa[0][i]));
-		asm volatile("vmovdqa %0,%%ymm1" : : "m" (p[1][i]));
-		asm volatile("vmovdqa %0,%%ymm3" : : "m" (pa[1][i]));
-		asm volatile("vpxor   %ymm2,%ymm0,%ymm0");
-		asm volatile("vpxor   %ymm3,%ymm1,%ymm1");
+		asm volatile ("vmovdqa %0,%%ymm0" : : "m" (p[0][i]));
+		asm volatile ("vmovdqa %0,%%ymm2" : : "m" (pa[0][i]));
+		asm volatile ("vmovdqa %0,%%ymm1" : : "m" (p[1][i]));
+		asm volatile ("vmovdqa %0,%%ymm3" : : "m" (pa[1][i]));
+		asm volatile ("vpxor   %ymm2,%ymm0,%ymm0");
+		asm volatile ("vpxor   %ymm3,%ymm1,%ymm1");
 
-		asm volatile("vpxor %ymm6,%ymm6,%ymm6");
+		asm volatile ("vpxor %ymm6,%ymm6,%ymm6");
 
-		asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[0]][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[0]][1][0]));
-		asm volatile("vpsrlw  $4,%ymm0,%ymm5");
-		asm volatile("vpand   %ymm7,%ymm0,%ymm4");
-		asm volatile("vpand   %ymm7,%ymm5,%ymm5");
-		asm volatile("vpshufb %ymm4,%ymm2,%ymm2");
-		asm volatile("vpshufb %ymm5,%ymm3,%ymm3");
-		asm volatile("vpxor   %ymm2,%ymm6,%ymm6");
-		asm volatile("vpxor   %ymm3,%ymm6,%ymm6");
+		asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[0]][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[0]][1][0]));
+		asm volatile ("vpsrlw  $4,%ymm0,%ymm5");
+		asm volatile ("vpand   %ymm7,%ymm0,%ymm4");
+		asm volatile ("vpand   %ymm7,%ymm5,%ymm5");
+		asm volatile ("vpshufb %ymm4,%ymm2,%ymm2");
+		asm volatile ("vpshufb %ymm5,%ymm3,%ymm3");
+		asm volatile ("vpxor   %ymm2,%ymm6,%ymm6");
+		asm volatile ("vpxor   %ymm3,%ymm6,%ymm6");
 
-		asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[1]][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[1]][1][0]));
-		asm volatile("vpsrlw  $4,%ymm1,%ymm5");
-		asm volatile("vpand   %ymm7,%ymm1,%ymm4");
-		asm volatile("vpand   %ymm7,%ymm5,%ymm5");
-		asm volatile("vpshufb %ymm4,%ymm2,%ymm2");
-		asm volatile("vpshufb %ymm5,%ymm3,%ymm3");
-		asm volatile("vpxor   %ymm2,%ymm6,%ymm6");
-		asm volatile("vpxor   %ymm3,%ymm6,%ymm6");
+		asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[1]][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[1]][1][0]));
+		asm volatile ("vpsrlw  $4,%ymm1,%ymm5");
+		asm volatile ("vpand   %ymm7,%ymm1,%ymm4");
+		asm volatile ("vpand   %ymm7,%ymm5,%ymm5");
+		asm volatile ("vpshufb %ymm4,%ymm2,%ymm2");
+		asm volatile ("vpshufb %ymm5,%ymm3,%ymm3");
+		asm volatile ("vpxor   %ymm2,%ymm6,%ymm6");
+		asm volatile ("vpxor   %ymm3,%ymm6,%ymm6");
 
-		asm volatile("vmovdqa %%ymm6,%0" : "=m" (pa[0][i]));
+		asm volatile ("vmovdqa %%ymm6,%0" : "=m" (pa[0][i]));
 
-		asm volatile("vpxor %ymm6,%ymm6,%ymm6");
+		asm volatile ("vpxor %ymm6,%ymm6,%ymm6");
 
-		asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[2]][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[2]][1][0]));
-		asm volatile("vpsrlw  $4,%ymm0,%ymm5");
-		asm volatile("vpand   %ymm7,%ymm0,%ymm4");
-		asm volatile("vpand   %ymm7,%ymm5,%ymm5");
-		asm volatile("vpshufb %ymm4,%ymm2,%ymm2");
-		asm volatile("vpshufb %ymm5,%ymm3,%ymm3");
-		asm volatile("vpxor   %ymm2,%ymm6,%ymm6");
-		asm volatile("vpxor   %ymm3,%ymm6,%ymm6");
+		asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[2]][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[2]][1][0]));
+		asm volatile ("vpsrlw  $4,%ymm0,%ymm5");
+		asm volatile ("vpand   %ymm7,%ymm0,%ymm4");
+		asm volatile ("vpand   %ymm7,%ymm5,%ymm5");
+		asm volatile ("vpshufb %ymm4,%ymm2,%ymm2");
+		asm volatile ("vpshufb %ymm5,%ymm3,%ymm3");
+		asm volatile ("vpxor   %ymm2,%ymm6,%ymm6");
+		asm volatile ("vpxor   %ymm3,%ymm6,%ymm6");
 
-		asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[3]][0][0]));
-		asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[3]][1][0]));
-		asm volatile("vpsrlw  $4,%ymm1,%ymm5");
-		asm volatile("vpand   %ymm7,%ymm1,%ymm4");
-		asm volatile("vpand   %ymm7,%ymm5,%ymm5");
-		asm volatile("vpshufb %ymm4,%ymm2,%ymm2");
-		asm volatile("vpshufb %ymm5,%ymm3,%ymm3");
-		asm volatile("vpxor   %ymm2,%ymm6,%ymm6");
-		asm volatile("vpxor   %ymm3,%ymm6,%ymm6");
+		asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[V[3]][0][0]));
+		asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[V[3]][1][0]));
+		asm volatile ("vpsrlw  $4,%ymm1,%ymm5");
+		asm volatile ("vpand   %ymm7,%ymm1,%ymm4");
+		asm volatile ("vpand   %ymm7,%ymm5,%ymm5");
+		asm volatile ("vpshufb %ymm4,%ymm2,%ymm2");
+		asm volatile ("vpshufb %ymm5,%ymm3,%ymm3");
+		asm volatile ("vpxor   %ymm2,%ymm6,%ymm6");
+		asm volatile ("vpxor   %ymm3,%ymm6,%ymm6");
 
-		asm volatile("vmovdqa %%ymm6,%0" : "=m" (pa[1][i]));
+		asm volatile ("vmovdqa %%ymm6,%0" : "=m" (pa[1][i]));
 	}
 
 	raid_asm_clobber_ymm8();
@@ -2450,41 +2450,41 @@ void raid_recX_avx2(int nr, int *id, int *ip, int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.low4[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm7" : : "m" (gfconst16.low4[0]));
 
 	for (i = 0; i < size; i += 32) {
 		uint8_t PD[RAID_PARITY_MAX][32] __aligned(32);
 
 		/* delta */
 		for (j = 0; j < N; ++j) {
-			asm volatile("vmovdqa %0,%%ymm0" : : "m" (p[j][i]));
-			asm volatile("vmovdqa %0,%%ymm1" : : "m" (pa[j][i]));
-			asm volatile("vpxor   %ymm1,%ymm0,%ymm0");
-			asm volatile("vmovdqa %%ymm0,%0" : "=m" (PD[j][0]));
+			asm volatile ("vmovdqa %0,%%ymm0" : : "m" (p[j][i]));
+			asm volatile ("vmovdqa %0,%%ymm1" : : "m" (pa[j][i]));
+			asm volatile ("vpxor   %ymm1,%ymm0,%ymm0");
+			asm volatile ("vmovdqa %%ymm0,%0" : "=m" (PD[j][0]));
 		}
 
 		/* reconstruct */
 		for (j = 0; j < N; ++j) {
-			asm volatile("vpxor %ymm0,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm1,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm0,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm1,%ymm1,%ymm1");
 
 			for (k = 0; k < N; ++k) {
 				uint8_t m = V[j*N+k];
 
-				asm volatile("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[m][0][0]));
-				asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[m][1][0]));
-				asm volatile("vmovdqa %0,%%ymm4" : : "m" (PD[k][0]));
-				asm volatile("vpsrlw  $4,%ymm4,%ymm5");
-				asm volatile("vpand   %ymm7,%ymm4,%ymm4");
-				asm volatile("vpand   %ymm7,%ymm5,%ymm5");
-				asm volatile("vpshufb %ymm4,%ymm2,%ymm2");
-				asm volatile("vpshufb %ymm5,%ymm3,%ymm3");
-				asm volatile("vpxor   %ymm2,%ymm0,%ymm0");
-				asm volatile("vpxor   %ymm3,%ymm1,%ymm1");
+				asm volatile ("vbroadcasti128 %0,%%ymm2" : : "m" (gfmulpshufb[m][0][0]));
+				asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfmulpshufb[m][1][0]));
+				asm volatile ("vmovdqa %0,%%ymm4" : : "m" (PD[k][0]));
+				asm volatile ("vpsrlw  $4,%ymm4,%ymm5");
+				asm volatile ("vpand   %ymm7,%ymm4,%ymm4");
+				asm volatile ("vpand   %ymm7,%ymm5,%ymm5");
+				asm volatile ("vpshufb %ymm4,%ymm2,%ymm2");
+				asm volatile ("vpshufb %ymm5,%ymm3,%ymm3");
+				asm volatile ("vpxor   %ymm2,%ymm0,%ymm0");
+				asm volatile ("vpxor   %ymm3,%ymm1,%ymm1");
 			}
 
-			asm volatile("vpxor %ymm1,%ymm0,%ymm0");
-			asm volatile("vmovdqa %%ymm0,%0" : "=m" (pa[j][i]));
+			asm volatile ("vpxor %ymm1,%ymm0,%ymm0");
+			asm volatile ("vmovdqa %%ymm0,%0" : "=m" (pa[j][i]));
 		}
 	}
 
diff --git a/raid/x86z.c b/raid/x86z.c
index 1da40e89..1e0fde6b 100644
--- a/raid/x86z.c
+++ b/raid/x86z.c
@@ -55,38 +55,38 @@ void raid_genz_sse2(int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfzconst16.poly[0]));
-	asm volatile("movdqa %0,%%xmm3" : : "m" (gfzconst16.half[0]));
-	asm volatile("movdqa %0,%%xmm6" : : "m" (gfzconst16.low7[0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfzconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm3" : : "m" (gfzconst16.half[0]));
+	asm volatile ("movdqa %0,%%xmm6" : : "m" (gfzconst16.low7[0]));
 
 	for (i = 0; i < size; i += 16) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (v[l][i]));
-		asm volatile("movdqa %xmm0,%xmm1");
-		asm volatile("movdqa %xmm0,%xmm2");
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (v[l][i]));
+		asm volatile ("movdqa %xmm0,%xmm1");
+		asm volatile ("movdqa %xmm0,%xmm2");
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("pxor %xmm4,%xmm4");
-			asm volatile("pcmpgtb %xmm1,%xmm4");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("pand %xmm7,%xmm4");
-			asm volatile("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm4,%xmm4");
+			asm volatile ("pcmpgtb %xmm1,%xmm4");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("pand %xmm7,%xmm4");
+			asm volatile ("pxor %xmm4,%xmm1");
 
-			asm volatile("movdqa %xmm2,%xmm4");
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("psllw $7,%xmm4");
-			asm volatile("psrlw $1,%xmm2");
-			asm volatile("pcmpgtb %xmm4,%xmm5");
-			asm volatile("pand %xmm6,%xmm2");
-			asm volatile("pand %xmm3,%xmm5");
-			asm volatile("pxor %xmm5,%xmm2");
+			asm volatile ("movdqa %xmm2,%xmm4");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("psllw $7,%xmm4");
+			asm volatile ("psrlw $1,%xmm2");
+			asm volatile ("pcmpgtb %xmm4,%xmm5");
+			asm volatile ("pand %xmm6,%xmm2");
+			asm volatile ("pand %xmm3,%xmm5");
+			asm volatile ("pxor %xmm5,%xmm2");
 
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm1");
-			asm volatile("pxor %xmm4,%xmm2");
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm4,%xmm2");
 		}
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
 	}
 
 	raid_asm_clobber_xmm8();
@@ -117,60 +117,60 @@ void raid_genz_sse2ext(int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("movdqa %0,%%xmm7" : : "m" (gfzconst16.poly[0]));
-	asm volatile("movdqa %0,%%xmm3" : : "m" (gfzconst16.half[0]));
-	asm volatile("movdqa %0,%%xmm11" : : "m" (gfzconst16.low7[0]));
+	asm volatile ("movdqa %0,%%xmm7" : : "m" (gfzconst16.poly[0]));
+	asm volatile ("movdqa %0,%%xmm3" : : "m" (gfzconst16.half[0]));
+	asm volatile ("movdqa %0,%%xmm11" : : "m" (gfzconst16.low7[0]));
 
 	for (i = 0; i < size; i += 32) {
-		asm volatile("movdqa %0,%%xmm0" : : "m" (v[l][i]));
-		asm volatile("movdqa %0,%%xmm8" : : "m" (v[l][i+16]));
-		asm volatile("movdqa %xmm0,%xmm1");
-		asm volatile("movdqa %xmm8,%xmm9");
-		asm volatile("movdqa %xmm0,%xmm2");
-		asm volatile("movdqa %xmm8,%xmm10");
+		asm volatile ("movdqa %0,%%xmm0" : : "m" (v[l][i]));
+		asm volatile ("movdqa %0,%%xmm8" : : "m" (v[l][i+16]));
+		asm volatile ("movdqa %xmm0,%xmm1");
+		asm volatile ("movdqa %xmm8,%xmm9");
+		asm volatile ("movdqa %xmm0,%xmm2");
+		asm volatile ("movdqa %xmm8,%xmm10");
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("movdqa %xmm2,%xmm6");
-			asm volatile("movdqa %xmm10,%xmm14");
-			asm volatile("pxor %xmm4,%xmm4");
-			asm volatile("pxor %xmm12,%xmm12");
-			asm volatile("pxor %xmm5,%xmm5");
-			asm volatile("pxor %xmm13,%xmm13");
-			asm volatile("psllw $7,%xmm6");
-			asm volatile("psllw $7,%xmm14");
-			asm volatile("psrlw $1,%xmm2");
-			asm volatile("psrlw $1,%xmm10");
-			asm volatile("pcmpgtb %xmm1,%xmm4");
-			asm volatile("pcmpgtb %xmm9,%xmm12");
-			asm volatile("pcmpgtb %xmm6,%xmm5");
-			asm volatile("pcmpgtb %xmm14,%xmm13");
-			asm volatile("paddb %xmm1,%xmm1");
-			asm volatile("paddb %xmm9,%xmm9");
-			asm volatile("pand %xmm11,%xmm2");
-			asm volatile("pand %xmm11,%xmm10");
-			asm volatile("pand %xmm7,%xmm4");
-			asm volatile("pand %xmm7,%xmm12");
-			asm volatile("pand %xmm3,%xmm5");
-			asm volatile("pand %xmm3,%xmm13");
-			asm volatile("pxor %xmm4,%xmm1");
-			asm volatile("pxor %xmm12,%xmm9");
-			asm volatile("pxor %xmm5,%xmm2");
-			asm volatile("pxor %xmm13,%xmm10");
+			asm volatile ("movdqa %xmm2,%xmm6");
+			asm volatile ("movdqa %xmm10,%xmm14");
+			asm volatile ("pxor %xmm4,%xmm4");
+			asm volatile ("pxor %xmm12,%xmm12");
+			asm volatile ("pxor %xmm5,%xmm5");
+			asm volatile ("pxor %xmm13,%xmm13");
+			asm volatile ("psllw $7,%xmm6");
+			asm volatile ("psllw $7,%xmm14");
+			asm volatile ("psrlw $1,%xmm2");
+			asm volatile ("psrlw $1,%xmm10");
+			asm volatile ("pcmpgtb %xmm1,%xmm4");
+			asm volatile ("pcmpgtb %xmm9,%xmm12");
+			asm volatile ("pcmpgtb %xmm6,%xmm5");
+			asm volatile ("pcmpgtb %xmm14,%xmm13");
+			asm volatile ("paddb %xmm1,%xmm1");
+			asm volatile ("paddb %xmm9,%xmm9");
+			asm volatile ("pand %xmm11,%xmm2");
+			asm volatile ("pand %xmm11,%xmm10");
+			asm volatile ("pand %xmm7,%xmm4");
+			asm volatile ("pand %xmm7,%xmm12");
+			asm volatile ("pand %xmm3,%xmm5");
+			asm volatile ("pand %xmm3,%xmm13");
+			asm volatile ("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm12,%xmm9");
+			asm volatile ("pxor %xmm5,%xmm2");
+			asm volatile ("pxor %xmm13,%xmm10");
 
-			asm volatile("movdqa %0,%%xmm4" : : "m" (v[d][i]));
-			asm volatile("movdqa %0,%%xmm12" : : "m" (v[d][i+16]));
-			asm volatile("pxor %xmm4,%xmm0");
-			asm volatile("pxor %xmm4,%xmm1");
-			asm volatile("pxor %xmm4,%xmm2");
-			asm volatile("pxor %xmm12,%xmm8");
-			asm volatile("pxor %xmm12,%xmm9");
-			asm volatile("pxor %xmm12,%xmm10");
+			asm volatile ("movdqa %0,%%xmm4" : : "m" (v[d][i]));
+			asm volatile ("movdqa %0,%%xmm12" : : "m" (v[d][i+16]));
+			asm volatile ("pxor %xmm4,%xmm0");
+			asm volatile ("pxor %xmm4,%xmm1");
+			asm volatile ("pxor %xmm4,%xmm2");
+			asm volatile ("pxor %xmm12,%xmm8");
+			asm volatile ("pxor %xmm12,%xmm9");
+			asm volatile ("pxor %xmm12,%xmm10");
 		}
-		asm volatile("movntdq %%xmm0,%0" : "=m" (p[i]));
-		asm volatile("movntdq %%xmm8,%0" : "=m" (p[i+16]));
-		asm volatile("movntdq %%xmm1,%0" : "=m" (q[i]));
-		asm volatile("movntdq %%xmm9,%0" : "=m" (q[i+16]));
-		asm volatile("movntdq %%xmm2,%0" : "=m" (r[i]));
-		asm volatile("movntdq %%xmm10,%0" : "=m" (r[i+16]));
+		asm volatile ("movntdq %%xmm0,%0" : "=m" (p[i]));
+		asm volatile ("movntdq %%xmm8,%0" : "=m" (p[i+16]));
+		asm volatile ("movntdq %%xmm1,%0" : "=m" (q[i]));
+		asm volatile ("movntdq %%xmm9,%0" : "=m" (q[i+16]));
+		asm volatile ("movntdq %%xmm2,%0" : "=m" (r[i]));
+		asm volatile ("movntdq %%xmm10,%0" : "=m" (r[i+16]));
 	}
 
 	raid_asm_clobber_xmm16();
@@ -201,55 +201,55 @@ void raid_genz_avx2ext(int nd, size_t size, void **vv)
 
 	raid_asm_begin();
 
-	asm volatile("vbroadcasti128 %0,%%ymm7" : : "m" (gfzconst16.poly[0]));
-	asm volatile("vbroadcasti128 %0,%%ymm3" : : "m" (gfzconst16.half[0]));
-	asm volatile("vbroadcasti128 %0,%%ymm11" : : "m" (gfzconst16.low7[0]));
-	asm volatile("vpxor %ymm15,%ymm15,%ymm15");
+	asm volatile ("vbroadcasti128 %0,%%ymm7" : : "m" (gfzconst16.poly[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm3" : : "m" (gfzconst16.half[0]));
+	asm volatile ("vbroadcasti128 %0,%%ymm11" : : "m" (gfzconst16.low7[0]));
+	asm volatile ("vpxor %ymm15,%ymm15,%ymm15");
 
 	for (i = 0; i < size; i += 64) {
-		asm volatile("vmovdqa %0,%%ymm0" : : "m" (v[l][i]));
-		asm volatile("vmovdqa %0,%%ymm8" : : "m" (v[l][i+32]));
-		asm volatile("vmovdqa %ymm0,%ymm1");
-		asm volatile("vmovdqa %ymm8,%ymm9");
-		asm volatile("vmovdqa %ymm0,%ymm2");
-		asm volatile("vmovdqa %ymm8,%ymm10");
+		asm volatile ("vmovdqa %0,%%ymm0" : : "m" (v[l][i]));
+		asm volatile ("vmovdqa %0,%%ymm8" : : "m" (v[l][i+32]));
+		asm volatile ("vmovdqa %ymm0,%ymm1");
+		asm volatile ("vmovdqa %ymm8,%ymm9");
+		asm volatile ("vmovdqa %ymm0,%ymm2");
+		asm volatile ("vmovdqa %ymm8,%ymm10");
 		for (d = l-1; d >= 0; --d) {
-			asm volatile("vpsllw $7,%ymm2,%ymm6");
-			asm volatile("vpsllw $7,%ymm10,%ymm14");
-			asm volatile("vpsrlw $1,%ymm2,%ymm2");
-			asm volatile("vpsrlw $1,%ymm10,%ymm10");
-			asm volatile("vpcmpgtb %ymm1,%ymm15,%ymm4");
-			asm volatile("vpcmpgtb %ymm9,%ymm15,%ymm12");
-			asm volatile("vpcmpgtb %ymm6,%ymm15,%ymm5");
-			asm volatile("vpcmpgtb %ymm14,%ymm15,%ymm13");
-			asm volatile("vpaddb %ymm1,%ymm1,%ymm1");
-			asm volatile("vpaddb %ymm9,%ymm9,%ymm9");
-			asm volatile("vpand %ymm11,%ymm2,%ymm2");
-			asm volatile("vpand %ymm11,%ymm10,%ymm10");
-			asm volatile("vpand %ymm7,%ymm4,%ymm4");
-			asm volatile("vpand %ymm7,%ymm12,%ymm12");
-			asm volatile("vpand %ymm3,%ymm5,%ymm5");
-			asm volatile("vpand %ymm3,%ymm13,%ymm13");
-			asm volatile("vpxor %ymm4,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm12,%ymm9,%ymm9");
-			asm volatile("vpxor %ymm5,%ymm2,%ymm2");
-			asm volatile("vpxor %ymm13,%ymm10,%ymm10");
+			asm volatile ("vpsllw $7,%ymm2,%ymm6");
+			asm volatile ("vpsllw $7,%ymm10,%ymm14");
+			asm volatile ("vpsrlw $1,%ymm2,%ymm2");
+			asm volatile ("vpsrlw $1,%ymm10,%ymm10");
+			asm volatile ("vpcmpgtb %ymm1,%ymm15,%ymm4");
+			asm volatile ("vpcmpgtb %ymm9,%ymm15,%ymm12");
+			asm volatile ("vpcmpgtb %ymm6,%ymm15,%ymm5");
+			asm volatile ("vpcmpgtb %ymm14,%ymm15,%ymm13");
+			asm volatile ("vpaddb %ymm1,%ymm1,%ymm1");
+			asm volatile ("vpaddb %ymm9,%ymm9,%ymm9");
+			asm volatile ("vpand %ymm11,%ymm2,%ymm2");
+			asm volatile ("vpand %ymm11,%ymm10,%ymm10");
+			asm volatile ("vpand %ymm7,%ymm4,%ymm4");
+			asm volatile ("vpand %ymm7,%ymm12,%ymm12");
+			asm volatile ("vpand %ymm3,%ymm5,%ymm5");
+			asm volatile ("vpand %ymm3,%ymm13,%ymm13");
+			asm volatile ("vpxor %ymm4,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm12,%ymm9,%ymm9");
+			asm volatile ("vpxor %ymm5,%ymm2,%ymm2");
+			asm volatile ("vpxor %ymm13,%ymm10,%ymm10");
 
-			asm volatile("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
-			asm volatile("vmovdqa %0,%%ymm12" : : "m" (v[d][i+32]));
-			asm volatile("vpxor %ymm4,%ymm0,%ymm0");
-			asm volatile("vpxor %ymm4,%ymm1,%ymm1");
-			asm volatile("vpxor %ymm4,%ymm2,%ymm2");
-			asm volatile("vpxor %ymm12,%ymm8,%ymm8");
-			asm volatile("vpxor %ymm12,%ymm9,%ymm9");
-			asm volatile("vpxor %ymm12,%ymm10,%ymm10");
+			asm volatile ("vmovdqa %0,%%ymm4" : : "m" (v[d][i]));
+			asm volatile ("vmovdqa %0,%%ymm12" : : "m" (v[d][i+32]));
+			asm volatile ("vpxor %ymm4,%ymm0,%ymm0");
+			asm volatile ("vpxor %ymm4,%ymm1,%ymm1");
+			asm volatile ("vpxor %ymm4,%ymm2,%ymm2");
+			asm volatile ("vpxor %ymm12,%ymm8,%ymm8");
+			asm volatile ("vpxor %ymm12,%ymm9,%ymm9");
+			asm volatile ("vpxor %ymm12,%ymm10,%ymm10");
 		}
-		asm volatile("vmovntdq %%ymm0,%0" : "=m" (p[i]));
-		asm volatile("vmovntdq %%ymm8,%0" : "=m" (p[i+32]));
-		asm volatile("vmovntdq %%ymm1,%0" : "=m" (q[i]));
-		asm volatile("vmovntdq %%ymm9,%0" : "=m" (q[i+32]));
-		asm volatile("vmovntdq %%ymm2,%0" : "=m" (r[i]));
-		asm volatile("vmovntdq %%ymm10,%0" : "=m" (r[i+32]));
+		asm volatile ("vmovntdq %%ymm0,%0" : "=m" (p[i]));
+		asm volatile ("vmovntdq %%ymm8,%0" : "=m" (p[i+32]));
+		asm volatile ("vmovntdq %%ymm1,%0" : "=m" (q[i]));
+		asm volatile ("vmovntdq %%ymm9,%0" : "=m" (q[i+32]));
+		asm volatile ("vmovntdq %%ymm2,%0" : "=m" (r[i]));
+		asm volatile ("vmovntdq %%ymm10,%0" : "=m" (r[i+32]));
 	}
 
 	raid_asm_clobber_ymm16();