erasure_code: fix modules using incorrect unsigned jump

Signed-off-by: Marcel Cornu <[email protected]>

Author: mdcornu

erasure_code/gf_2vect_dot_prod_avx512_gfni.asm

@@ -177,7 +177,7 @@ func(gf_2vect_dot_prod_avx512_gfni)
 	mov	dest1, [dest1]
 
 	cmp	len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64:
 

erasure_code/gf_3vect_dot_prod_avx2_gfni.asm

@@ -300,7 +300,7 @@ func(gf_3vect_dot_prod_avx2_gfni)
     mov     dest3, [dest + 2*8]
 
     cmp     len, 64
-    jb      .len_lt_64
+    jl      .len_lt_64
 
 .loop64:
     ENCODE_64B_3
@@ -312,7 +312,7 @@ func(gf_3vect_dot_prod_avx2_gfni)
 
 .len_lt_64:
     cmp     len, 32
-    jb      .len_lt_32
+    jl      .len_lt_32
 
     ENCODE_32B_3
 

erasure_code/gf_3vect_dot_prod_avx512_gfni.asm

@@ -193,7 +193,7 @@ func(gf_3vect_dot_prod_avx512_gfni)
 	mov	dest1, [dest1]
 
 	cmp	len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64:
 

erasure_code/gf_3vect_mad_avx2_gfni.asm

@@ -241,7 +241,7 @@ func(gf_3vect_mad_avx2_gfni)
         mov     dest1, [dest1]
 
         cmp     len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64:
         ENCODE_64B_3            ;; loop on 64 bytes at a time
@@ -253,7 +253,7 @@ func(gf_3vect_mad_avx2_gfni)
 
 .len_lt_64:
         cmp     len, 32
-        jb      .len_lt_32
+        jl      .len_lt_32
 
         ENCODE_32B_3            ;; encode next 32 bytes
 

erasure_code/gf_4vect_dot_prod_avx512_gfni.asm

@@ -221,7 +221,7 @@ func(gf_4vect_dot_prod_avx512_gfni)
 	mov	dest1, [dest1]
 
 	cmp	len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64:
 

erasure_code/gf_4vect_mad_avx2_gfni.asm

@@ -213,7 +213,7 @@ func(gf_4vect_mad_avx2_gfni)
         mov     dest1, [dest1]
 
         cmp     len, 32
-        jb      .len_lt_32
+        jl      .len_lt_32
 
 .loop32:
         ENCODE_32B_4            ;; loop on 32 bytes at a time

erasure_code/gf_5vect_dot_prod_avx512_gfni.asm

@@ -242,7 +242,7 @@ func(gf_5vect_dot_prod_avx512_gfni)
 	mov	dest1, [dest1]
 
 	cmp	len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64:
 

erasure_code/gf_5vect_mad_avx2_gfni.asm

@@ -239,7 +239,7 @@ func(gf_5vect_mad_avx2_gfni)
         mov     dest1, [dest1]
 
         cmp     len, 32
-        jb      .len_lt_32
+        jl      .len_lt_32
 
 .loop32:
         ENCODE_32B_5            ;; loop on 32 bytes at a time

erasure_code/gf_6vect_dot_prod_avx512_gfni.asm

@@ -260,7 +260,7 @@ func(gf_6vect_dot_prod_avx512_gfni)
 	mov	dest5, [dest1 + 4*8]      ;dest1 and dest6 are calculated later
 
 	cmp	len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64:
 

erasure_code/gf_vect_dot_prod_avx512_gfni.asm

@@ -158,7 +158,7 @@ func(gf_vect_dot_prod_avx512_gfni)
 	shl	vec, 3		;vec *= 8. Make vec_i count by 8
 
 	cmp	len, 64
-        jb      .len_lt_64
+        jl      .len_lt_64
 
 .loop64: