(mpn_sqr): New name for mpn_sqr_n.

37 files changed, 107 insertions, 160 deletions

diff --git a/ChangeLog b/ChangeLog
index 2ccddc484..65f7e38d6 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -6,6 +6,8 @@
 
 2009-12-27  Torbjorn Granlund  <tege@gmplib.org>
 
+	* (mpn_sqr): New name for mpn_sqr_n.  Many files affected.
+
 	* tune/tuneup.c (tune_mullo): Up step_factor for MULLO_MUL_N_THRESHOLD.
 	(tune_invertappr, tune_invert, tune_binvert): Let max_size default.
 
diff --git a/NEWS b/NEWS
index 7e67f2648..37e7c3cbb 100644
--- a/NEWS
+++ b/NEWS
@@ -40,7 +40,7 @@ Changes between GMP version 4.3.X and 4.4.0
     average, where Q is the quotient.
 
   Features:
-  * New mpn functions: mpn_sqr_n, mpn_and_n, mpn_ior_n, mpn_xor_n,
+  * New mpn functions: mpn_sqr, mpn_and_n, mpn_ior_n, mpn_xor_n,
     mpn_nand_n, mpn_nior_n, mpn_xnor_n, mpn_andn_n, mpn_iorn_n.
   * Support for fat binaries for 64-bit x86 processors.
   * New type, mp_bitcnt_t for bignum bit counts.
diff --git a/configure.in b/configure.in
index cbe14f16c..6137efcae 100644
--- a/configure.in
+++ b/configure.in
@@ -2495,7 +2495,7 @@ gmp_mpn_functions="$extra_functions					   \
   submul_1 lshift rshift dive_1 diveby3 divis divrem divrem_1 divrem_2     \
   fib2_ui mod_1 mod_34lsub1 mode1o pre_divrem_1 pre_mod_1 dump		   \
   mod_1_1 mod_1_2 mod_1_3 mod_1_4					   \
-  mul mul_fft mul_n sqr_n mul_basecase sqr_basecase nussbaumer_mul	   \
+  mul mul_fft mul_n sqr mul_basecase sqr_basecase nussbaumer_mul	   \
   random random2 pow_1							   \
   rootrem sqrtrem get_str set_str scan0 scan1 popcount hamdist cmp	   \
   perfsqr perfpow							   \
diff --git a/doc/gmp.texi b/doc/gmp.texi
index 234ac9243..e77e2289a 100644
--- a/doc/gmp.texi
+++ b/doc/gmp.texi
@@ -5175,7 +5175,7 @@ The destination has to have space for 2*@var{n} limbs, even if the product's
 most significant limb is zero.  No overlap is permitted between the
 destination and either source.
 
-If the two input operands are the same, use @code{mpn_sqr_n}.
+If the two input operands are the same, use @code{mpn_sqr}.
 @end deftypefun
 
 @deftypefun mp_limb_t mpn_mul (mp_limb_t *@var{rp}, const mp_limb_t *@var{s1p}, mp_size_t @var{s1n}, const mp_limb_t *@var{s2p}, mp_size_t @var{s2n})
@@ -5190,7 +5190,7 @@ destination and either source.
 This function requires that @var{s1n} is greater than or equal to @var{s2n}.
 @end deftypefun
 
-@deftypefun void mpn_sqr_n (mp_limb_t *@var{rp}, const mp_limb_t *@var{s1p}, mp_size_t @var{n})
+@deftypefun void mpn_sqr (mp_limb_t *@var{rp}, const mp_limb_t *@var{s1p}, mp_size_t @var{n})
 Compute the square of @{@var{s1p}, @var{n}@} and write the 2*@var{n}-limb
 result to @var{rp}.
 
diff --git a/gmp-h.in b/gmp-h.in
index 31410c1df..af79f66a9 100644
--- a/gmp-h.in
+++ b/gmp-h.in
@@ -1568,6 +1568,9 @@ __GMP_DECLSPEC mp_limb_t mpn_mul_1 __GMP_PROTO ((mp_ptr, mp_srcptr, mp_size_t, m
 #define mpn_mul_n __MPN(mul_n)
 __GMP_DECLSPEC void mpn_mul_n __GMP_PROTO ((mp_ptr, mp_srcptr, mp_srcptr, mp_size_t));
 
+#define mpn_sqr __MPN(sqr)
+__GMP_DECLSPEC void mpn_sqr __GMP_PROTO ((mp_ptr, mp_srcptr, mp_size_t));
+
 #define mpn_neg_n __MPN(neg_n)
 #if __GMP_INLINE_PROTOTYPES || defined (__GMP_FORCE_mpn_neg_n)
 __GMP_DECLSPEC mp_limb_t mpn_neg_n __GMP_PROTO ((mp_ptr, mp_srcptr, mp_size_t));
diff --git a/gmp-impl.h b/gmp-impl.h
index f4bf1174b..eb0c90cf1 100644
--- a/gmp-impl.h
+++ b/gmp-impl.h
@@ -908,8 +908,8 @@ __GMP_DECLSPEC void mpn_mullo_n __GMP_PROTO ((mp_ptr, mp_srcptr, mp_srcptr, mp_s
 #define mpn_mullo_basecase __MPN(mullo_basecase)
 __GMP_DECLSPEC void mpn_mullo_basecase __GMP_PROTO ((mp_ptr, mp_srcptr, mp_srcptr, mp_size_t));
 
-#define mpn_sqr_n __MPN(sqr_n)
-__GMP_DECLSPEC void mpn_sqr_n __GMP_PROTO ((mp_ptr, mp_srcptr, mp_size_t));
+#define mpn_sqr __MPN(sqr)
+__GMP_DECLSPEC void mpn_sqr __GMP_PROTO ((mp_ptr, mp_srcptr, mp_size_t));
 
 #ifndef mpn_sqr_basecase  /* if not done with cpuvec in a fat binary */
 #define mpn_sqr_basecase __MPN(sqr_basecase)
diff --git a/mpf/get_str.c b/mpf/get_str.c
index 30b12e7c2..447bfdbb9 100644
--- a/mpf/get_str.c
+++ b/mpf/get_str.c
@@ -67,7 +67,7 @@ mpn_pow_1_highpart (mp_ptr rp, mp_size_t *ignp,
   count_leading_zeros (cnt, exp);
   for (i = GMP_LIMB_BITS - cnt - 2; i >= 0; i--)
     {
-      mpn_sqr_n (tp, rp + off, rn);
+      mpn_sqr (tp, rp + off, rn);
       rn = 2 * rn;
       rn -= tp[rn - 1] == 0;
       ign <<= 1;
diff --git a/mpf/set_str.c b/mpf/set_str.c
index ce8a6bac3..01a175fa6 100644
--- a/mpf/set_str.c
+++ b/mpf/set_str.c
@@ -72,7 +72,7 @@ mpn_pow_1_highpart (mp_ptr rp, mp_size_t *ignp,
   count_leading_zeros (cnt, exp);
   for (i = GMP_LIMB_BITS - cnt - 2; i >= 0; i--)
     {
-      mpn_sqr_n (tp, rp + off, rn);
+      mpn_sqr (tp, rp + off, rn);
       rn = 2 * rn;
       rn -= tp[rn - 1] == 0;
       ign <<= 1;
diff --git a/mpn/asm-defs.m4 b/mpn/asm-defs.m4
index f66ac66a7..9ad92c182 100644
--- a/mpn/asm-defs.m4
+++ b/mpn/asm-defs.m4
@@ -1347,8 +1347,6 @@ define_mpn(hamdist)
 define_mpn(invert_limb)
 define_mpn(ior_n)
 define_mpn(iorn_n)
-define_mpn(kara_mul_n)
-define_mpn(kara_sqr_n)
 define_mpn(lshift)
 define_mpn(lshiftc)
 define_mpn(mod_1_1p)
@@ -1410,8 +1408,6 @@ define_mpn(sub_n)
 define_mpn(sub_nc)
 define_mpn(submul_1)
 define_mpn(submul_1c)
-define_mpn(toom3_mul_n)
-define_mpn(toom3_sqr_n)
 define_mpn(umul_ppmm)
 define_mpn(umul_ppmm_r)
 define_mpn(udiv_qrnnd)
diff --git a/mpn/generic/fib2_ui.c b/mpn/generic/fib2_ui.c
index 437375353..ddf93faed 100644
--- a/mpn/generic/fib2_ui.c
+++ b/mpn/generic/fib2_ui.c
@@ -105,8 +105,8 @@ mpn_fib2_ui (mp_ptr fp, mp_ptr f1p, unsigned long int n)
 	  /* f1p[size-1] might be zero, but this occurs rarely, so it's not
 	     worth bothering checking for it */
 	  ASSERT (alloc >= 2*size);
-	  mpn_sqr_n (xp, fp,  size);
-	  mpn_sqr_n (fp, f1p, size);
+	  mpn_sqr (xp, fp,  size);
+	  mpn_sqr (fp, f1p, size);
 	  size *= 2;
 
 	  /* Shrink if possible.  Since fp was normalized there'll be at
diff --git a/mpn/generic/get_str.c b/mpn/generic/get_str.c
index a00fb805f..ac4fb52a9 100644
--- a/mpn/generic/get_str.c
+++ b/mpn/generic/get_str.c
@@ -473,7 +473,7 @@ mpn_get_str (unsigned char *str, int base, mp_ptr up, mp_size_t un)
 
 	ASSERT_ALWAYS (powtab_mem_ptr < powtab_mem + mpn_dc_get_str_powtab_alloc (un));
 
-	mpn_sqr_n (t, p, n);
+	mpn_sqr (t, p, n);
 
 	digits_in_base *= 2;
 	n *= 2;  n -= t[n - 1] == 0;
diff --git a/mpn/generic/mul.c b/mpn/generic/mul.c
index 02134e573..055f78fde 100644
--- a/mpn/generic/mul.c
+++ b/mpn/generic/mul.c
@@ -84,7 +84,7 @@ mpn_mul (mp_ptr prodp,
   if (un == vn)
     {
       if (up == vp)
-	mpn_sqr_n (prodp, up, un);
+	mpn_sqr (prodp, up, un);
       else
 	mpn_mul_n (prodp, up, vp, un);
     }
diff --git a/mpn/generic/mul_fft.c b/mpn/generic/mul_fft.c
index f10185eea..1315b0cf5 100644
--- a/mpn/generic/mul_fft.c
+++ b/mpn/generic/mul_fft.c
@@ -529,7 +529,7 @@ mpn_fft_mul_modF_K (mp_ptr *ap, mp_ptr *bp, mp_size_t n, int K)
 	  a = *ap++;
 	  b = *bp++;
 	  if (sqr)
-	    mpn_sqr_n (tp, a, n);
+	    mpn_sqr (tp, a, n);
 	  else
 	    mpn_mul_n (tp, b, a, n);
 	  if (a[n] != 0)
diff --git a/mpn/generic/perfpow.c b/mpn/generic/perfpow.c
index 54143a790..ccd2b6891 100644
--- a/mpn/generic/perfpow.c
+++ b/mpn/generic/perfpow.c
@@ -405,7 +405,7 @@ mpn_perfect_power_p (mp_srcptr np, mp_size_t nn)
 	  exp = 1;
 	  while (2 * pn - 1 <= ncn)
 	    {
-	      mpn_sqr_n (next, prev, pn);
+	      mpn_sqr (next, prev, pn);
 	      xn = 2 * pn;
 	      xn -= (next[xn - 1] == 0);
 
diff --git a/mpn/generic/pow_1.c b/mpn/generic/pow_1.c
index 88d66ea0e..d379836e3 100644
--- a/mpn/generic/pow_1.c
+++ b/mpn/generic/pow_1.c
@@ -71,7 +71,7 @@ mpn_pow_1 (mp_ptr rp, mp_srcptr bp, mp_size_t bn, mp_limb_t exp, mp_ptr tp)
       if ((cnt & 1) != 0)
 	MP_PTR_SWAP (rp, tp);
 
-      mpn_sqr_n (rp, bp, bn);
+      mpn_sqr (rp, bp, bn);
       rn = 2 * bn; rn -= rp[rn - 1] == 0;
 
       for (i = GMP_LIMB_BITS - cnt - 1;;)
@@ -86,7 +86,7 @@ mpn_pow_1 (mp_ptr rp, mp_srcptr bp, mp_size_t bn, mp_limb_t exp, mp_ptr tp)
 	  if (--i == 0)
 	    break;
 
-	  mpn_sqr_n (tp, rp, rn);
+	  mpn_sqr (tp, rp, rn);
 	  rn = 2 * rn; rn -= tp[rn - 1] == 0;
 	  MP_PTR_SWAP (rp, tp);
 	}
@@ -96,7 +96,7 @@ mpn_pow_1 (mp_ptr rp, mp_srcptr bp, mp_size_t bn, mp_limb_t exp, mp_ptr tp)
       if (((par ^ cnt) & 1) == 0)
 	MP_PTR_SWAP (rp, tp);
 
-      mpn_sqr_n (rp, bp, bn);
+      mpn_sqr (rp, bp, bn);
       rn = 2 * bn; rn -= rp[rn - 1] == 0;
 
       for (i = GMP_LIMB_BITS - cnt - 1;;)
@@ -111,7 +111,7 @@ mpn_pow_1 (mp_ptr rp, mp_srcptr bp, mp_size_t bn, mp_limb_t exp, mp_ptr tp)
 	  if (--i == 0)
 	    break;
 
-	  mpn_sqr_n (tp, rp, rn);
+	  mpn_sqr (tp, rp, rn);
 	  rn = 2 * rn; rn -= tp[rn - 1] == 0;
 	  MP_PTR_SWAP (rp, tp);
 	}
diff --git a/mpn/generic/powlo.c b/mpn/generic/powlo.c
index 23990ada8..7eb64540d 100644
--- a/mpn/generic/powlo.c
+++ b/mpn/generic/powlo.c
@@ -95,7 +95,7 @@ mpn_powlo (mp_ptr rp, mp_srcptr bp,
   b2p = tp + 2*n;
 
   /* Store b^2 in b2.  */
-  mpn_sqr_n (tp, bp, n);	/* FIXME: Use "mpn_sqrlo" */
+  mpn_sqr (tp, bp, n);	/* FIXME: Use "mpn_sqrlo" */
   MPN_COPY (b2p, tp, n);
 
   /* Precompute odd powers of b and put them in the temporary area at pp.  */
@@ -122,7 +122,7 @@ mpn_powlo (mp_ptr rp, mp_srcptr bp,
     {
       while (getbit (ep, ebi) == 0)
 	{
-	  mpn_sqr_n (tp, rp, n);	/* FIXME: Use "mpn_sqrlo" */
+	  mpn_sqr (tp, rp, n);	/* FIXME: Use "mpn_sqrlo" */
 	  MPN_COPY (rp, tp, n);
 	  ebi--;
 	  if (ebi == 0)
@@ -149,7 +149,7 @@ mpn_powlo (mp_ptr rp, mp_srcptr bp,
 
       do
 	{
-	  mpn_sqr_n (tp, rp, n);
+	  mpn_sqr (tp, rp, n);
 	  MPN_COPY (rp, tp, n);
 	  this_windowsize--;
 	}
diff --git a/mpn/generic/powm.c b/mpn/generic/powm.c
index 712de6c94..7010357f1 100644
--- a/mpn/generic/powm.c
+++ b/mpn/generic/powm.c
@@ -166,7 +166,7 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
 	 until the result is greater than the mod argument.  */
       for (;;)
 	{
-	  mpn_sqr_n (tp, this_pp, tn);
+	  mpn_sqr (tp, this_pp, tn);
 	  tn = tn * 2 - 1,  tn += tp[tn] != 0;
 	  if (getbit (ep, ebi) != 0)
 	    mpn_mul (..., tp, tn, bp, bn);
@@ -212,7 +212,7 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   b2p = tp + 2*n;
 
   /* Store b^2 in b2.  */
-  mpn_sqr_n (tp, this_pp, n);
+  mpn_sqr (tp, this_pp, n);
 #if WANT_REDC_2
   if (BELOW_THRESHOLD (n, REDC_1_TO_REDC_2_THRESHOLD))
     mpn_redc_1 (b2p, tp, mp, n, mip[0]);
@@ -260,7 +260,7 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
     {									\
       while (getbit (ep, ebi) == 0)					\
 	{								\
-	  MPN_SQR_N (tp, rp, n);					\
+	  MPN_SQR (tp, rp, n);						\
 	  MPN_REDUCE (rp, tp, mp, n, mip);				\
 	  ebi--;							\
 	  if (ebi == 0)							\
@@ -288,7 +288,7 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
 									\
       do								\
 	{								\
-	  MPN_SQR_N (tp, rp, n);					\
+	  MPN_SQR (tp, rp, n);						\
 	  MPN_REDUCE (rp, tp, mp, n, mip);				\
 	  this_windowsize--;						\
 	}								\
@@ -304,20 +304,20 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   if (BELOW_THRESHOLD (n, REDC_1_TO_REDC_2_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_basecase (r,a,n,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_basecase (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr_basecase (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_1 (rp, tp, mp, n, mip[0])
       INNERLOOP;
     }
   else if (BELOW_THRESHOLD (n, MUL_TOOM22_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_basecase (r,a,n,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_basecase (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr_basecase (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_2 (rp, tp, mp, n, mip)
       INNERLOOP;
     }
@@ -325,20 +325,20 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   if (BELOW_THRESHOLD (n, MUL_TOOM22_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_basecase (r,a,n,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_basecase (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr_basecase (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_1 (rp, tp, mp, n, mip[0])
       INNERLOOP;
     }
   else if (BELOW_THRESHOLD (n, REDC_1_TO_REDC_2_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_n (r,a,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_n (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_1 (rp, tp, mp, n, mip[0])
       INNERLOOP;
     }
@@ -346,20 +346,20 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   else if (BELOW_THRESHOLD (n, REDC_2_TO_REDC_N_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_n (r,a,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_n (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_2 (rp, tp, mp, n, mip)
       INNERLOOP;
     }
   else
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_n (r,a,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_n (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_n (rp, tp, mp, n, mip)
       INNERLOOP;
     }
@@ -369,20 +369,20 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   if (BELOW_THRESHOLD (n, REDC_1_TO_REDC_N_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_basecase (r,a,n,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_basecase (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr_basecase (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_1 (rp, tp, mp, n, mip[0])
       INNERLOOP;
     }
   else if (BELOW_THRESHOLD (n, MUL_TOOM22_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_basecase (r,a,n,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_basecase (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr_basecase (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_n (rp, tp, mp, n, mip)
       INNERLOOP;
     }
@@ -390,20 +390,20 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   if (BELOW_THRESHOLD (n, MUL_TOOM22_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_basecase (r,a,n,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_basecase (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr_basecase (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_1 (rp, tp, mp, n, mip[0])
       INNERLOOP;
     }
   else if (BELOW_THRESHOLD (n, REDC_1_TO_REDC_N_THRESHOLD))
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_n (r,a,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_n (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_1 (rp, tp, mp, n, mip[0])
       INNERLOOP;
     }
@@ -411,10 +411,10 @@ mpn_powm (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
   else
     {
 #undef MPN_MUL_N
-#undef MPN_SQR_N
+#undef MPN_SQR
 #undef MPN_REDUCE
 #define MPN_MUL_N(r,a,b,n)		mpn_mul_n (r,a,b,n)
-#define MPN_SQR_N(r,a,n)		mpn_sqr_n (r,a,n)
+#define MPN_SQR(r,a,n)			mpn_sqr (r,a,n)
 #define MPN_REDUCE(rp,tp,mp,n,mip)	mpn_redc_n (rp, tp, mp, n, mip)
       INNERLOOP;
     }
diff --git a/mpn/generic/powm_sec.c b/mpn/generic/powm_sec.c
index e94fd8c57..315ae6e5e 100644
--- a/mpn/generic/powm_sec.c
+++ b/mpn/generic/powm_sec.c
@@ -98,12 +98,12 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
 #ifndef SQR_BASECASE_MAX
 /* If SQR_BASECASE_MAX is now not defined, use mpn_sqr_basecase for any operand
    size.  */
-#define mpn_local_sqr_n(rp,up,n,tp) mpn_sqr_basecase(rp,up,n)
+#define mpn_local_sqr(rp,up,n,tp) mpn_sqr_basecase(rp,up,n)
 #else
 /* Define our own squaring function, which uses mpn_sqr_basecase for its
    allowed sizes, but its own code for larger sizes.  */
 static void
-mpn_local_sqr_n (mp_ptr rp, mp_srcptr up, mp_size_t n, mp_ptr tp)
+mpn_local_sqr (mp_ptr rp, mp_srcptr up, mp_size_t n, mp_ptr tp)
 {
   mp_size_t i;
 
@@ -277,7 +277,7 @@ mpn_powm_sec (mp_ptr rp, mp_srcptr bp, mp_size_t bn,
 
       do
 	{
-	  mpn_local_sqr_n (tp, rp, n, tp + 2 * n);
+	  mpn_local_sqr (tp, rp, n, tp + 2 * n);
 	  mpn_redc_1_sec (rp, tp, mp, n, minv);
 	  this_windowsize--;
 	}
diff --git a/mpn/generic/remove.c b/mpn/generic/remove.c
index 1fe3cbc0f..427a46fbd 100644
--- a/mpn/generic/remove.c
+++ b/mpn/generic/remove.c
@@ -104,7 +104,7 @@ mpn_remove (mp_ptr wp, mp_size_t *wn,
       if (nn > qn)
 	break;			/* next power would be overlarge */
 
-      mpn_sqr_n (np, pp, pn);
+      mpn_sqr (np, pp, pn);
       nn += np[nn] != 0;
       pp = np;
       pn = nn;
diff --git a/mpn/generic/set_str.c b/mpn/generic/set_str.c
index db916e8c5..83f5ac550 100644
--- a/mpn/generic/set_str.c
+++ b/mpn/generic/set_str.c
@@ -169,7 +169,7 @@ mpn_set_str_compute_powtab (powers_t *powtab, mp_ptr powtab_mem, mp_size_t un, i
 
       ASSERT_ALWAYS (powtab_mem_ptr < powtab_mem + mpn_dc_set_str_powtab_alloc (un));
 
-      mpn_sqr_n (t, p, n);
+      mpn_sqr (t, p, n);
       n = 2 * n - 1; n += t[n] != 0;
       digits_in_base *= 2;
 #if 1
diff --git a/mpn/generic/sqr_n.c b/mpn/generic/sqr.c
index e492b5bb2..504dbfb2d 100644
--- a/mpn/generic/sqr_n.c
+++ b/mpn/generic/sqr.c
@@ -1,4 +1,4 @@
-/* mpn_sqr_n -- square natural numbers.
+/* mpn_sqr -- square natural numbers.
 
 Copyright 1991, 1993, 1994, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003,
 2005, 2008, 2009 Free Software Foundation, Inc.
@@ -23,7 +23,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
 #include "longlong.h"
 
 void
-mpn_sqr_n (mp_ptr p, mp_srcptr a, mp_size_t n)
+mpn_sqr (mp_ptr p, mp_srcptr a, mp_size_t n)
 {
   ASSERT (n >= 1);
   ASSERT (! MPN_OVERLAP_P (p, 2 * n, a, n));
diff --git a/mpn/generic/sqrmod_bnm1.c b/mpn/generic/sqrmod_bnm1.c
index c1b192942..d07933c7d 100644
--- a/mpn/generic/sqrmod_bnm1.c
+++ b/mpn/generic/sqrmod_bnm1.c
@@ -40,7 +40,7 @@ mpn_bc_sqrmod_bnm1 (mp_ptr rp, mp_srcptr ap, mp_size_t rn, mp_ptr tp)
 
   ASSERT (0 < rn);
 
-  mpn_sqr_n (tp, ap, rn);
+  mpn_sqr (tp, ap, rn);
   cy = mpn_add_n (rp, tp, tp + rn, rn);
   /* If cy == 1, then the value of rp is at most B^rn - 2, so there can
    * be no overflow when adding in the carry. */
@@ -59,7 +59,7 @@ mpn_bc_sqrmod_bnp1 (mp_ptr rp, mp_srcptr ap, mp_size_t rn, mp_ptr tp)
 
   ASSERT (0 < rn);
 
-  mpn_sqr_n (tp, ap, rn + 1);
+  mpn_sqr (tp, ap, rn + 1);
   ASSERT (tp[2*rn+1] == 0);
   ASSERT (tp[2*rn] < GMP_NUMB_MAX);
   cy = tp[2*rn] + mpn_sub_n (rp, tp, tp+rn, rn);
@@ -94,13 +94,13 @@ mpn_sqrmod_bnm1 (mp_ptr rp, mp_size_t rn, mp_srcptr ap, mp_size_t an, mp_ptr tp)
 	{
 	  if (UNLIKELY (2*an <= rn))
 	    {
-	      mpn_sqr_n (rp, ap, an);
+	      mpn_sqr (rp, ap, an);
 	      MPN_ZERO (rp + 2*an, rn - 2*an);
 	    }
 	  else
 	    {
 	      mp_limb_t cy;
-	      mpn_sqr_n (tp, ap, an);
+	      mpn_sqr (tp, ap, an);
 	      cy = mpn_add (rp, tp, rn, tp + rn, 2*an - rn);
 	      MPN_INCR_U (rp, rn, cy);
 	    }
diff --git a/mpn/generic/sqrtrem.c b/mpn/generic/sqrtrem.c
index ac878c508..87852f40d 100644
--- a/mpn/generic/sqrtrem.c
+++ b/mpn/generic/sqrtrem.c
@@ -239,7 +239,7 @@ mpn_dc_sqrtrem (mp_ptr sp, mp_ptr np, mp_size_t n)
       q >>= 1;
       if (c != 0)
 	c = mpn_add_n (np + l, np + l, sp + l, h);
-      mpn_sqr_n (np + n, sp, l);
+      mpn_sqr (np + n, sp, l);
       b = q + mpn_sub_n (np, np, np + n, 2 * l);
       c -= (l == h) ? b : mpn_sub_1 (np + 2 * l, np + 2 * l, 1, (mp_limb_t) b);
       q = mpn_add_1 (sp + l, sp + l, h, q);
diff --git a/mpn/generic/toom2_sqr.c b/mpn/generic/toom2_sqr.c
index 4c27aeb4e..08fe7999a 100644
--- a/mpn/generic/toom2_sqr.c
+++ b/mpn/generic/toom2_sqr.c
@@ -45,7 +45,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
   (SQR_TOOM3_THRESHOLD >= 2 * SQR_TOOM2_THRESHOLD)
 #endif
 
-#define TOOM2_SQR_N_REC(p, a, n, ws)					\
+#define TOOM2_SQR_REC(p, a, n, ws)					\
   do {									\
     if (! MAYBE_sqr_toom2						\
 	|| BELOW_THRESHOLD (n, SQR_TOOM2_THRESHOLD))			\
@@ -104,13 +104,13 @@ mpn_toom2_sqr (mp_ptr pp,
 #define scratch_out	scratch + 2 * n
 
   /* vm1, 2n limbs */
-  TOOM2_SQR_N_REC (vm1, asm1, n, scratch_out);
+  TOOM2_SQR_REC (vm1, asm1, n, scratch_out);
 
   /* vinf, s+s limbs */
-  TOOM2_SQR_N_REC (vinf, a1, s, scratch_out);
+  TOOM2_SQR_REC (vinf, a1, s, scratch_out);
 
   /* v0, 2n limbs */
-  TOOM2_SQR_N_REC (v0, ap, n, scratch_out);
+  TOOM2_SQR_REC (v0, ap, n, scratch_out);
 
   /* H(v0) + L(vinf) */
   cy = mpn_add_n (pp + 2 * n, v0 + n, vinf, n);
diff --git a/mpn/generic/toom3_sqr.c b/mpn/generic/toom3_sqr.c
index a85e546fe..f44fff451 100644
--- a/mpn/generic/toom3_sqr.c
+++ b/mpn/generic/toom3_sqr.c
@@ -51,7 +51,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
   (SQR_TOOM4_THRESHOLD >= 3 * SQR_TOOM3_THRESHOLD)
 #endif
 
-#define TOOM3_SQR_N_REC(p, a, n, ws)					\
+#define TOOM3_SQR_REC(p, a, n, ws)					\
   do {									\
     if (MAYBE_sqr_basecase						\
 	&& BELOW_THRESHOLD (n, SQR_TOOM2_THRESHOLD))			\
@@ -154,7 +154,7 @@ mpn_toom3_sqr (mp_ptr pp,
 
   /* vm1, 2n+1 limbs */
 #ifdef SMALLER_RECURSION
-  TOOM3_SQR_N_REC (vm1, asm1, n, scratch_out);
+  TOOM3_SQR_REC (vm1, asm1, n, scratch_out);
   cy = 0;
   if (asm1[n] != 0)
     cy = asm1[n] + mpn_add_n (vm1 + n, vm1 + n, asm1, n);
@@ -162,18 +162,18 @@ mpn_toom3_sqr (mp_ptr pp,
     cy += mpn_add_n (vm1 + n, vm1 + n, asm1, n);
   vm1[2 * n] = cy;
 #else
-  TOOM3_SQR_N_REC (vm1, asm1, n + 1, scratch_out);
+  TOOM3_SQR_REC (vm1, asm1, n + 1, scratch_out);
 #endif
 
-  TOOM3_SQR_N_REC (v2, as2, n + 1, scratch_out);	/* v2, 2n+1 limbs */
+  TOOM3_SQR_REC (v2, as2, n + 1, scratch_out);	/* v2, 2n+1 limbs */
 
-  TOOM3_SQR_N_REC (vinf, a2, s, scratch_out);		/* vinf, s+s limbs */
+  TOOM3_SQR_REC (vinf, a2, s, scratch_out);	/* vinf, s+s limbs */
 
   vinf0 = vinf[0];				/* v1 overlaps with this */
 
 #ifdef SMALLER_RECURSION
   /* v1, 2n+1 limbs */
-  TOOM3_SQR_N_REC (v1, as1, n, scratch_out);
+  TOOM3_SQR_REC (v1, as1, n, scratch_out);
   if (as1[n] == 1)
     {
       cy = as1[n] + mpn_add_n (v1 + n, v1 + n, as1, n);
@@ -203,11 +203,11 @@ mpn_toom3_sqr (mp_ptr pp,
   v1[2 * n] = cy;
 #else
   cy = vinf[1];
-  TOOM3_SQR_N_REC (v1, as1, n + 1, scratch_out);
+  TOOM3_SQR_REC (v1, as1, n + 1, scratch_out);
   vinf[1] = cy;
 #endif
 
-  TOOM3_SQR_N_REC (v0, ap, n, scratch_out);	/* v0, 2n limbs */
+  TOOM3_SQR_REC (v0, ap, n, scratch_out);	/* v0, 2n limbs */
 
   mpn_toom_interpolate_5pts (pp, v2, vm1, n, s + s, 0, vinf0);
 }
diff --git a/mpn/generic/toom4_sqr.c b/mpn/generic/toom4_sqr.c
index e188ee0ae..68aec769c 100644
--- a/mpn/generic/toom4_sqr.c
+++ b/mpn/generic/toom4_sqr.c
@@ -55,7 +55,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
   (SQR_FFT_THRESHOLD >= 4 * SQR_TOOM4_THRESHOLD)
 #endif
 
-#define TOOM4_SQR_N_REC(p, a, n, ws)					\
+#define TOOM4_SQR_REC(p, a, n, ws)					\
   do {									\
     if (MAYBE_sqr_basecase						\
 	&& BELOW_THRESHOLD (n, SQR_TOOM2_THRESHOLD))			\
@@ -111,8 +111,8 @@ mpn_toom4_sqr (mp_ptr pp,
   /* Compute apx = a0 + 2 a1 + 4 a2 + 8 a3 and amx = a0 - 2 a1 + 4 a2 - 8 a3.  */
   mpn_toom_eval_dgr3_pm2 (apx, amx, ap, n, s, tp);
 
-  TOOM4_SQR_N_REC (v2, apx, n + 1, tp);	/* v2,  2n+1 limbs */
-  TOOM4_SQR_N_REC (vm2, amx, n + 1, tp);	/* vm2,  2n+1 limbs */
+  TOOM4_SQR_REC (v2, apx, n + 1, tp);	/* v2,  2n+1 limbs */
+  TOOM4_SQR_REC (vm2, amx, n + 1, tp);	/* vm2,  2n+1 limbs */
 
   /* Compute apx = 8 a0 + 4 a1 + 2 a2 + a3 = (((2*a0 + a1) * 2 + a2) * 2 + a3 */
 #if HAVE_NATIVE_mpn_addlsh1_n
@@ -138,16 +138,16 @@ mpn_toom4_sqr (mp_ptr pp,
 
   ASSERT (apx[n] < 15);
 
-  TOOM4_SQR_N_REC (vh, apx, n + 1, tp);	/* vh,  2n+1 limbs */
+  TOOM4_SQR_REC (vh, apx, n + 1, tp);	/* vh,  2n+1 limbs */
 
   /* Compute apx = a0 + a1 + a2 + a3 and amx = a0 - a1 + a2 - a3.  */
   mpn_toom_eval_dgr3_pm1 (apx, amx, ap, n, s, tp);
 
-  TOOM4_SQR_N_REC (v1, apx, n + 1, tp);	/* v1,  2n+1 limbs */
-  TOOM4_SQR_N_REC (vm1, amx, n + 1, tp);	/* vm1,  2n+1 limbs */
+  TOOM4_SQR_REC (v1, apx, n + 1, tp);	/* v1,  2n+1 limbs */
+  TOOM4_SQR_REC (vm1, amx, n + 1, tp);	/* vm1,  2n+1 limbs */
 
-  TOOM4_SQR_N_REC (v0, a0, n, tp);
-  TOOM4_SQR_N_REC (vinf, a3, s, tp);	/* vinf, 2s limbs */
+  TOOM4_SQR_REC (v0, a0, n, tp);
+  TOOM4_SQR_REC (vinf, a3, s, tp);	/* vinf, 2s limbs */
 
   mpn_toom_interpolate_7pts (pp, n, 0, vm2, vm1, v2, vh, 2*s, tp);
 }
diff --git a/mpn/x86/fat/gmp-mparam.h b/mpn/x86/fat/gmp-mparam.h
index 051a87d19..45680ede4 100644
--- a/mpn/x86/fat/gmp-mparam.h
+++ b/mpn/x86/fat/gmp-mparam.h
@@ -35,7 +35,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
 #define USE_PREINV_DIVREM_1   1
 
 /* mpn_sqr_basecase is faster than mpn_mul_basecase at all sizes, no need
-   for mpn_sqr_n to call the latter.  */
+   for mpn_sqr to call the latter.  */
 #define SQR_BASECASE_THRESHOLD 0
 
 /* Sensible fallbacks for these, when not taken from a cpu-specific
diff --git a/mpn/x86_64/fat/gmp-mparam.h b/mpn/x86_64/fat/gmp-mparam.h
index 7e71f715b..6e744c064 100644
--- a/mpn/x86_64/fat/gmp-mparam.h
+++ b/mpn/x86_64/fat/gmp-mparam.h
@@ -35,7 +35,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
 #define USE_PREINV_DIVREM_1   1
 
 /* mpn_sqr_basecase is faster than mpn_mul_basecase at all sizes, no need
-   for mpn_sqr_n to call the latter.  */
+   for mpn_sqr to call the latter.  */
 #define SQR_BASECASE_THRESHOLD 0
 
 /* Sensible fallbacks for these, when not taken from a cpu-specific
diff --git a/mpz/lucnum_ui.c b/mpz/lucnum_ui.c
index 1215a04b4..1fb8ec86c 100644
--- a/mpz/lucnum_ui.c
+++ b/mpz/lucnum_ui.c
@@ -167,7 +167,7 @@ mpz_lucnum_ui (mpz_ptr ln, unsigned long n)
       TRACE (printf ("  zeros=%d\n", zeros));
 
       ASSERT (xalloc >= 2*lsize);
-      mpn_sqr_n (xp, lp, lsize);
+      mpn_sqr (xp, lp, lsize);
       lsize *= 2;
       lsize -= (xp[lsize-1] == 0);
 
diff --git a/mpz/n_pow_ui.c b/mpz/n_pow_ui.c
index c1d5e902d..6d527c36c 100644
--- a/mpz/n_pow_ui.c
+++ b/mpz/n_pow_ui.c
@@ -59,7 +59,7 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
    from mpn_mul_1 or mpn_mul_2 in the bignum powering.  It's felt that doing
    so would be more complicated than it's worth, and could well end up being
    a slowdown for small e.  For big e on the other hand the algorithm is
-   dominated by mpn_sqr_n so there wouldn't much of a saving.  The current
+   dominated by mpn_sqr so there wouldn't much of a saving.  The current
    code can be viewed as simply doing the first few steps of the powering in
    a single or double limb where possible.
 
@@ -79,10 +79,10 @@ along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.  */
 /* The following are for convenience, they update the size and check the
    alloc.  */
 
-#define MPN_SQR_N(dst, alloc, src, size)        \
+#define MPN_SQR(dst, alloc, src, size)          \
   do {                                          \
     ASSERT (2*(size) <= (alloc));               \
-    mpn_sqr_n (dst, src, size);                 \
+    mpn_sqr (dst, src, size);                   \
     (size) *= 2;                                \
     (size) -= ((dst)[(size)-1] == 0);           \
   } while (0)
@@ -435,7 +435,7 @@ mpz_n_pow_ui (mpz_ptr r, mp_srcptr bp, mp_size_t bsize, unsigned long int e)
                              i, e, rsize, ralloc, talloc);
                      mpn_trace ("r", rp, rsize));
 
-              MPN_SQR_N (tp, talloc, rp, rsize);
+              MPN_SQR (tp, talloc, rp, rsize);
               SWAP_RP_TP;
               if ((e & (1L << i)) != 0)
                 MPN_MUL_2 (rp, rsize, ralloc, mult);
@@ -467,7 +467,7 @@ mpz_n_pow_ui (mpz_ptr r, mp_srcptr bp, mp_size_t bsize, unsigned long int e)
                              i, e, rsize, ralloc, talloc);
                      mpn_trace ("r", rp, rsize));
 
-              MPN_SQR_N (tp, talloc, rp, rsize);
+              MPN_SQR (tp, talloc, rp, rsize);
               SWAP_RP_TP;
               if ((e & (1L << i)) != 0)
                 MPN_MUL_1 (rp, rsize, ralloc, blimb);
@@ -496,7 +496,7 @@ mpz_n_pow_ui (mpz_ptr r, mp_srcptr bp, mp_size_t bsize, unsigned long int e)
                              i, e, rsize, ralloc, talloc);
                      mpn_trace ("r", rp, rsize));
 
-              MPN_SQR_N (tp, talloc, rp, rsize);
+              MPN_SQR (tp, talloc, rp, rsize);
               SWAP_RP_TP;
               if ((e & (1L << i)) != 0)
                 {
diff --git a/mpz/powm_ui.c b/mpz/powm_ui.c
index 6200f3333..64615d107 100644
--- a/mpz/powm_ui.c
+++ b/mpz/powm_ui.c
@@ -126,7 +126,7 @@ mpz_powm_ui (mpz_ptr r, mpz_srcptr b, unsigned long int el, mpz_srcptr m)
 
   while (c != 0)
     {
-      mpn_sqr_n (tp, xp, xn);
+      mpn_sqr (tp, xp, xn);
       tn = 2 * xn; tn -= tp[tn - 1] == 0;
       if (tn < mn)
 	{
diff --git a/tests/devel/try.c b/tests/devel/try.c
index 516915f53..3a4f1dc28 100644
--- a/tests/devel/try.c
+++ b/tests/devel/try.c
@@ -1562,7 +1562,7 @@ const struct choice_t choice_array[] = {
 
   { TRY(mpn_mul),    TYPE_MUL_MN },
   { TRY(mpn_mul_n),  TYPE_MUL_N },
-  { TRY(mpn_sqr_n),  TYPE_SQR },
+  { TRY(mpn_sqr),    TYPE_SQR },
 
   { TRY_FUNFUN(umul_ppmm), TYPE_UMUL_PPMM, 2 },
 #if HAVE_NATIVE_mpn_umul_ppmm
diff --git a/tune/Makefile.am b/tune/Makefile.am
index 2d5b05420..022aa7c17 100644
--- a/tune/Makefile.am
+++ b/tune/Makefile.am
@@ -125,7 +125,7 @@ TUNE_MPN_SRCS = $(TUNE_MPN_SRCS_BASIC) divrem_1.c mod_1.c
 TUNE_MPN_SRCS_BASIC = bdiv_q.c bdiv_qr.c				\
   dcpi1_div_qr.c dcpi1_divappr_q.c dcpi1_bdiv_qr.c dcpi1_bdiv_q.c	\
   invertappr.c invert.c binvert.c divrem_2.c gcd.c gcdext.c		\
-  get_str.c set_str.c matrix22_mul.c hgcd.c mul_n.c sqr_n.c		\
+  get_str.c set_str.c matrix22_mul.c hgcd.c mul_n.c sqr.c		\
   mullo_n.c mul_fft.c mul.c tdiv_qr.c mulmod_bnm1.c sqrmod_bnm1.c	\
   nussbaumer_mul.c toom6h_mul.c toom8h_mul.c toom6_sqr.c toom8_sqr.c	\
   toom22_mul.c toom2_sqr.c toom33_mul.c toom3_sqr.c toom44_mul.c toom4_sqr.c
@@ -152,52 +152,3 @@ sqr_asm.asm: $(top_builddir)/mpn/sqr_basecase.asm
 
 
 include ../mpn/Makeasm.am
-
-
-# "mk" is multiplication in the karatsuba range
-# "st" is squaring in the toom-cook range, etc
-# "g" forms produce graphs
-
-mk:
-	./speed -s 5-40 -c mpn_mul_basecase mpn_kara_mul_n
-
-MTS = -s 50-150 -c
-mt:
-	./speed $(MTS) mpn_kara_mul_n mpn_toom3_mul_n
-mtg:
-	./speed $(MTS) -P mtg mpn_kara_mul_n mpn_toom3_mul_n
-
-sk:
-	./speed -s 5-40 -c mpn_sqr_basecase mpn_kara_sqr_n
-
-STS = -s 50-150 -c
-st:
-	./speed $(STS) mpn_kara_sqr_n mpn_toom3_sqr_n
-stg:
-	./speed $(STS) -P stg mpn_kara_sqr_n mpn_toom3_sqr_n
-
-fib:
-	./speed -s 40-60 -c mpz_fib_ui
-fibg:
-	./speed -s 10-300 -P fibg mpz_fib_ui
-
-
-gcd:
-	./speed -s 1-20 -c mpn_gcd
-
-udiv:
-	./speed -s 1 -c udiv_qrnnd udiv_qrnnd_preinv udiv_qrnnd_preinv2norm invert_limb udiv_qrnnd_c
-
-divn:
-	./speed -s 1-30 -c mpn_divrem_1_div.-1 mpn_divrem_1_inv.-1
-divun:
-	./speed -s 1-30 -c mpn_divrem_1_div.12345 mpn_divrem_1_inv.12345
-modn:
-	./speed -s 1-30 -c mpn_mod_1_div.-1 mpn_mod_1_inv.-1
-modun:
-	./speed -s 1-30 -c mpn_mod_1_div.12345 mpn_mod_1_inv.12345
-
-
-graph:
-	./speed -s 1-5000 -f 1.02 -P graph mpn_mul_n mpn_sqr
-	gnuplot graph.gnuplot
diff --git a/tune/common.c b/tune/common.c
index 63a30b39b..d44e62cb2 100644
--- a/tune/common.c
+++ b/tune/common.c
@@ -1003,9 +1003,9 @@ speed_mpn_mul_n (struct speed_params *s)
   SPEED_ROUTINE_MPN_MUL_N (mpn_mul_n);
 }
 double
-speed_mpn_sqr_n (struct speed_params *s)
+speed_mpn_sqr (struct speed_params *s)
 {
-  SPEED_ROUTINE_MPN_SQR (mpn_sqr_n);
+  SPEED_ROUTINE_MPN_SQR (mpn_sqr);
 }
 double
 speed_mpn_mul_n_sqr (struct speed_params *s)
diff --git a/tune/speed.c b/tune/speed.c
index 179bb29b8..a9e3508c5 100644
--- a/tune/speed.c
+++ b/tune/speed.c
@@ -294,7 +294,7 @@ const struct routine_t {
 #endif
 
   { "mpn_mul_n",         speed_mpn_mul_n            },
-  { "mpn_sqr_n",         speed_mpn_sqr_n            },
+  { "mpn_sqr",           speed_mpn_sqr              },
 
   { "mpn_toom2_sqr",     speed_mpn_toom2_sqr        },
   { "mpn_toom3_sqr",     speed_mpn_toom3_sqr        },
diff --git a/tune/speed.h b/tune/speed.h
index 1ed5aeec1..a807c7c56 100644
--- a/tune/speed.h
+++ b/tune/speed.h
@@ -273,7 +273,7 @@ double speed_mpn_dc_set_str __GMP_PROTO ((struct speed_params *s));
 double speed_mpn_set_str_pre __GMP_PROTO ((struct speed_params *s));
 double speed_mpn_sqr_basecase __GMP_PROTO ((struct speed_params *s));
 double speed_mpn_sqr_diagonal __GMP_PROTO ((struct speed_params *s));
-double speed_mpn_sqr_n __GMP_PROTO ((struct speed_params *s));
+double speed_mpn_sqr __GMP_PROTO ((struct speed_params *s));
 double speed_mpn_sqrtrem __GMP_PROTO ((struct speed_params *s));
 double speed_mpn_rootrem __GMP_PROTO ((struct speed_params *s));
 double speed_mpn_sub_n __GMP_PROTO ((struct speed_params *s));
@@ -445,11 +445,6 @@ mp_limb_t mpn_sb_divrem_mn_inv __GMP_PROTO ((mp_ptr, mp_ptr, mp_size_t, mp_srcpt
 mp_size_t mpn_set_str_basecase __GMP_PROTO ((mp_ptr, const unsigned char *, size_t, int));
 void mpn_pre_set_str __GMP_PROTO ((mp_ptr, unsigned char *, size_t, powers_t *, mp_ptr));
 
-void mpn_toom3_mul_n_open __GMP_PROTO ((mp_ptr, mp_srcptr, mp_srcptr, mp_size_t, mp_ptr));
-void mpn_toom3_sqr_n_open __GMP_PROTO ((mp_ptr, mp_srcptr, mp_size_t, mp_ptr));
-void mpn_toom3_mul_n_mpn __GMP_PROTO ((mp_ptr, mp_srcptr, mp_srcptr, mp_size_t, mp_ptr));
-void mpn_toom3_sqr_n_mpn __GMP_PROTO((mp_ptr, mp_srcptr, mp_size_t, mp_ptr));
-
 void mpz_powm_mod __GMP_PROTO ((mpz_ptr, mpz_srcptr, mpz_srcptr, mpz_srcptr));
 void mpz_powm_redc __GMP_PROTO ((mpz_ptr, mpz_srcptr, mpz_srcptr, mpz_srcptr));
 
diff --git a/tune/tuneup.c b/tune/tuneup.c
index 6eeaa3d7f..a9fc89e69 100644
--- a/tune/tuneup.c
+++ b/tune/tuneup.c
@@ -992,7 +992,7 @@ tune_sqrmod_bnm1 (void)
    just for that.  Start karatsuba from 4 same as MUL above.  */
 
 void
-tune_sqr_n (void)
+tune_sqr (void)
 {
   /* disabled until tuned */
   SQR_FFT_THRESHOLD = MP_SIZE_T_MAX;
@@ -1006,7 +1006,7 @@ tune_sqr_n (void)
     {
       static struct param_t  param;
       param.name = "SQR_BASECASE_THRESHOLD";
-      param.function = speed_mpn_sqr_n;
+      param.function = speed_mpn_sqr;
       param.min_size = 3;
       param.min_is_always = 1;
       param.max_size = TUNE_SQR_TOOM2_MAX;
@@ -1017,7 +1017,7 @@ tune_sqr_n (void)
   {
     static struct param_t  param;
     param.name = "SQR_TOOM2_THRESHOLD";
-    param.function = speed_mpn_sqr_n;
+    param.function = speed_mpn_sqr;
     param.min_size = MAX (4, MPN_TOOM2_SQR_MINSIZE);
     param.max_size = TUNE_SQR_TOOM2_MAX;
     param.noprint = 1;
@@ -1029,7 +1029,7 @@ tune_sqr_n (void)
         /* Karatsuba becomes faster than mul_basecase before
            sqr_basecase does.  Arrange for the expression
            "BELOW_THRESHOLD (un, SQR_TOOM2_THRESHOLD))" which
-           selects mpn_sqr_basecase in mpn_sqr_n to be false, by setting
+           selects mpn_sqr_basecase in mpn_sqr to be false, by setting
            SQR_TOOM2_THRESHOLD to zero, making
            SQR_BASECASE_THRESHOLD the toom2 threshold.  */
 
@@ -1053,7 +1053,7 @@ tune_sqr_n (void)
     static struct param_t  param;
     mp_size_t toom3_start = MAX (sqr_toom2_threshold, sqr_basecase_threshold);
 
-    param.function = speed_mpn_sqr_n;
+    param.function = speed_mpn_sqr;
 
     param.name = "SQR_TOOM3_THRESHOLD";
     param.min_size = MAX (toom3_start, MPN_TOOM3_SQR_MINSIZE);
@@ -1927,7 +1927,7 @@ tune_fft_sqr (void)
   param.first_size          = SQR_TOOM3_THRESHOLD / 2;
   param.max_size            = option_fft_max_size;
   param.function            = speed_mpn_mul_fft_sqr;
-  param.mul_function        = speed_mpn_sqr_n;
+  param.mul_function        = speed_mpn_sqr;
   param.sqr = 0;
   fft (&param);
 }
@@ -2012,7 +2012,7 @@ all (void)
   tune_mul ();
   printf("\n");
 
-  tune_sqr_n ();
+  tune_sqr ();
   printf("\n");
 
   tune_mulmod_bnm1 ();