diff -Naurd mpfr-3.1.6-a/PATCHES mpfr-3.1.6-b/PATCHES
--- mpfr-3.1.6-a/PATCHES	2017-10-26 13:55:51.168013439 +0000
+++ mpfr-3.1.6-b/PATCHES	2017-10-26 13:55:51.236013121 +0000
@@ -0,0 +1 @@
+mpfr_get
diff -Naurd mpfr-3.1.6-a/VERSION mpfr-3.1.6-b/VERSION
--- mpfr-3.1.6-a/VERSION	2017-09-07 11:36:44.000000000 +0000
+++ mpfr-3.1.6-b/VERSION	2017-10-26 13:55:51.236013121 +0000
@@ -1 +1 @@
-3.1.6
+3.1.6-p1
diff -Naurd mpfr-3.1.6-a/src/get_ld.c mpfr-3.1.6-b/src/get_ld.c
--- mpfr-3.1.6-a/src/get_ld.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/get_ld.c	2017-10-26 13:55:51.208013252 +0000
@@ -41,6 +41,9 @@
       mpfr_exp_t sh; /* exponent shift, so that x/2^sh is in the double range */
       mpfr_t y, z;
       int sign;
+      MPFR_SAVE_EXPO_DECL (expo);
+
+      MPFR_SAVE_EXPO_MARK (expo);
 
       /* first round x to the target long double precision, so that
          all subsequent operations are exact (this avoids double rounding
@@ -103,6 +106,7 @@
         }
       if (sign < 0)
         r = -r;
+      MPFR_SAVE_EXPO_FREE (expo);
       return r;
     }
 }
diff -Naurd mpfr-3.1.6-a/src/get_si.c mpfr-3.1.6-b/src/get_si.c
--- mpfr-3.1.6-a/src/get_si.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/get_si.c	2017-10-26 13:55:51.208013252 +0000
@@ -28,6 +28,7 @@
   mpfr_prec_t prec;
   long s;
   mpfr_t x;
+  MPFR_SAVE_EXPO_DECL (expo);
 
   if (MPFR_UNLIKELY (!mpfr_fits_slong_p (f, rnd)))
     {
@@ -39,14 +40,22 @@
   if (MPFR_IS_ZERO (f))
     return (long) 0;
 
-  /* determine prec of long */
-  for (s = LONG_MIN, prec = 0; s != 0; s /= 2, prec++)
+  /* Determine the precision of long. |LONG_MIN| may have one more bit
+     as an integer, but in this case, this is a power of 2, thus fits
+     in a precision-prec floating-point number. */
+  for (s = LONG_MAX, prec = 0; s != 0; s /= 2, prec++)
     { }
 
+  MPFR_SAVE_EXPO_MARK (expo);
+
   /* first round to prec bits */
   mpfr_init2 (x, prec);
   mpfr_rint (x, f, rnd);
 
+  /* The flags from mpfr_rint are the wanted ones. In particular,
+     it sets the inexact flag when necessary. */
+  MPFR_SAVE_EXPO_UPDATE_FLAGS (expo, __gmpfr_flags);
+
   /* warning: if x=0, taking its exponent is illegal */
   if (MPFR_UNLIKELY (MPFR_IS_ZERO(x)))
     s = 0;
@@ -65,5 +74,7 @@
 
   mpfr_clear (x);
 
+  MPFR_SAVE_EXPO_FREE (expo);
+
   return s;
 }
diff -Naurd mpfr-3.1.6-a/src/get_sj.c mpfr-3.1.6-b/src/get_sj.c
--- mpfr-3.1.6-a/src/get_sj.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/get_sj.c	2017-10-26 13:55:51.208013252 +0000
@@ -35,6 +35,7 @@
   intmax_t r;
   mpfr_prec_t prec;
   mpfr_t x;
+  MPFR_SAVE_EXPO_DECL (expo);
 
   if (MPFR_UNLIKELY (!mpfr_fits_intmax_p (f, rnd)))
     {
@@ -46,20 +47,24 @@
   if (MPFR_IS_ZERO (f))
      return (intmax_t) 0;
 
-  /* determine the precision of intmax_t */
-  for (r = MPFR_INTMAX_MIN, prec = 0; r != 0; r /= 2, prec++)
+  /* Determine the precision of intmax_t. |INTMAX_MIN| may have one
+     more bit as an integer, but in this case, this is a power of 2,
+     thus fits in a precision-prec floating-point number. */
+  for (r = MPFR_INTMAX_MAX, prec = 0; r != 0; r /= 2, prec++)
     { }
-  /* Note: though INTMAX_MAX would have been sufficient for the conversion,
-     we chose INTMAX_MIN so that INTMAX_MIN - 1 is always representable in
-     precision prec; this is useful to detect overflows in MPFR_RNDZ (will
-     be needed later). */
 
-  /* Now, r = 0. */
+  MPFR_ASSERTD (r == 0);
+
+  MPFR_SAVE_EXPO_MARK (expo);
 
   mpfr_init2 (x, prec);
   mpfr_rint (x, f, rnd);
   MPFR_ASSERTN (MPFR_IS_FP (x));
 
+  /* The flags from mpfr_rint are the wanted ones. In particular,
+     it sets the inexact flag when necessary. */
+  MPFR_SAVE_EXPO_UPDATE_FLAGS (expo, __gmpfr_flags);
+
   if (MPFR_NOTZERO (x))
     {
       mp_limb_t *xp;
@@ -67,15 +72,15 @@
 
       xp = MPFR_MANT (x);
       sh = MPFR_GET_EXP (x);
-      MPFR_ASSERTN ((mpfr_prec_t) sh <= prec);
+      MPFR_ASSERTN ((mpfr_prec_t) sh <= prec + 1);
       if (MPFR_INTMAX_MIN + MPFR_INTMAX_MAX != 0
-          && MPFR_UNLIKELY ((mpfr_prec_t) sh == prec))
+          && MPFR_UNLIKELY ((mpfr_prec_t) sh > prec))
         {
           /* 2's complement and x <= INTMAX_MIN: in the case mp_limb_t
              has the same size as intmax_t, we cannot use the code in
              the for loop since the operations would be performed in
              unsigned arithmetic. */
-          MPFR_ASSERTN (MPFR_IS_NEG (x) && (mpfr_powerof2_raw (x)));
+          MPFR_ASSERTN (MPFR_IS_NEG (x) && mpfr_powerof2_raw (x));
           r = MPFR_INTMAX_MIN;
         }
       else if (MPFR_IS_POS (x))
@@ -117,6 +122,8 @@
 
   mpfr_clear (x);
 
+  MPFR_SAVE_EXPO_FREE (expo);
+
   return r;
 }
 
diff -Naurd mpfr-3.1.6-a/src/get_ui.c mpfr-3.1.6-b/src/get_ui.c
--- mpfr-3.1.6-a/src/get_ui.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/get_ui.c	2017-10-26 13:55:51.208013252 +0000
@@ -30,6 +30,7 @@
   mpfr_t x;
   mp_size_t n;
   mpfr_exp_t exp;
+  MPFR_SAVE_EXPO_DECL (expo);
 
   if (MPFR_UNLIKELY (!mpfr_fits_ulong_p (f, rnd)))
     {
@@ -44,10 +45,16 @@
   for (s = ULONG_MAX, prec = 0; s != 0; s /= 2, prec ++)
     { }
 
+  MPFR_SAVE_EXPO_MARK (expo);
+
   /* first round to prec bits */
   mpfr_init2 (x, prec);
   mpfr_rint (x, f, rnd);
 
+  /* The flags from mpfr_rint are the wanted ones. In particular,
+     it sets the inexact flag when necessary. */
+  MPFR_SAVE_EXPO_UPDATE_FLAGS (expo, __gmpfr_flags);
+
   /* warning: if x=0, taking its exponent is illegal */
   if (MPFR_IS_ZERO(x))
     s = 0;
@@ -61,5 +68,7 @@
 
   mpfr_clear (x);
 
+  MPFR_SAVE_EXPO_FREE (expo);
+
   return s;
 }
diff -Naurd mpfr-3.1.6-a/src/get_uj.c mpfr-3.1.6-b/src/get_uj.c
--- mpfr-3.1.6-a/src/get_uj.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/get_uj.c	2017-10-26 13:55:51.208013252 +0000
@@ -35,6 +35,7 @@
   uintmax_t r;
   mpfr_prec_t prec;
   mpfr_t x;
+  MPFR_SAVE_EXPO_DECL (expo);
 
   if (MPFR_UNLIKELY (!mpfr_fits_uintmax_p (f, rnd)))
     {
@@ -50,12 +51,18 @@
   for (r = MPFR_UINTMAX_MAX, prec = 0; r != 0; r /= 2, prec++)
     { }
 
-  /* Now, r = 0. */
+  MPFR_ASSERTD (r == 0);
+
+  MPFR_SAVE_EXPO_MARK (expo);
 
   mpfr_init2 (x, prec);
   mpfr_rint (x, f, rnd);
   MPFR_ASSERTN (MPFR_IS_FP (x));
 
+  /* The flags from mpfr_rint are the wanted ones. In particular,
+     it sets the inexact flag when necessary. */
+  MPFR_SAVE_EXPO_UPDATE_FLAGS (expo, __gmpfr_flags);
+
   if (MPFR_NOTZERO (x))
     {
       mp_limb_t *xp;
@@ -76,6 +83,8 @@
 
   mpfr_clear (x);
 
+  MPFR_SAVE_EXPO_FREE (expo);
+
   return r;
 }
 
diff -Naurd mpfr-3.1.6-a/src/get_z.c mpfr-3.1.6-b/src/get_z.c
--- mpfr-3.1.6-a/src/get_z.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/get_z.c	2017-10-26 13:55:51.208013252 +0000
@@ -29,6 +29,7 @@
   int inex;
   mpfr_t r;
   mpfr_exp_t exp;
+  MPFR_SAVE_EXPO_DECL (expo);
 
   if (MPFR_UNLIKELY (MPFR_IS_SINGULAR (f)))
     {
@@ -41,6 +42,8 @@
       return 0;
     }
 
+  MPFR_SAVE_EXPO_MARK (expo);
+
   exp = MPFR_GET_EXP (f);
   /* if exp <= 0, then |f|<1, thus |o(f)|<=1 */
   MPFR_ASSERTN (exp < 0 || exp <= MPFR_PREC_MAX);
@@ -50,6 +53,11 @@
   MPFR_ASSERTN (inex != 1 && inex != -1); /* integral part of f is
                                              representable in r */
   MPFR_ASSERTN (MPFR_IS_FP (r));
+
+  /* The flags from mpfr_rint are the wanted ones. In particular,
+     it sets the inexact flag when necessary. */
+  MPFR_SAVE_EXPO_UPDATE_FLAGS (expo, __gmpfr_flags);
+
   exp = mpfr_get_z_2exp (z, r);
   if (exp >= 0)
     mpz_mul_2exp (z, z, exp);
@@ -57,5 +65,7 @@
     mpz_fdiv_q_2exp (z, z, -exp);
   mpfr_clear (r);
 
+  MPFR_SAVE_EXPO_FREE (expo);
+
   return inex;
 }
diff -Naurd mpfr-3.1.6-a/src/mpfr.h mpfr-3.1.6-b/src/mpfr.h
--- mpfr-3.1.6-a/src/mpfr.h	2017-09-07 11:36:44.000000000 +0000
+++ mpfr-3.1.6-b/src/mpfr.h	2017-10-26 13:55:51.232013138 +0000
@@ -27,7 +27,7 @@
 #define MPFR_VERSION_MAJOR 3
 #define MPFR_VERSION_MINOR 1
 #define MPFR_VERSION_PATCHLEVEL 6
-#define MPFR_VERSION_STRING "3.1.6"
+#define MPFR_VERSION_STRING "3.1.6-p1"
 
 /* Macros dealing with MPFR VERSION */
 #define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.6-a/src/version.c mpfr-3.1.6-b/src/version.c
--- mpfr-3.1.6-a/src/version.c	2017-09-07 11:36:44.000000000 +0000
+++ mpfr-3.1.6-b/src/version.c	2017-10-26 13:55:51.232013138 +0000
@@ -25,5 +25,5 @@
 const char *
 mpfr_get_version (void)
 {
-  return "3.1.6";
+  return "3.1.6-p1";
 }
diff -Naurd mpfr-3.1.6-a/PATCHES mpfr-3.1.6-b/PATCHES
--- mpfr-3.1.6-a/PATCHES	2017-12-15 12:37:44.074256548 +0000
+++ mpfr-3.1.6-b/PATCHES	2017-12-15 12:37:44.146256304 +0000
@@ -0,0 +1 @@
+root
diff -Naurd mpfr-3.1.6-a/VERSION mpfr-3.1.6-b/VERSION
--- mpfr-3.1.6-a/VERSION	2017-10-26 13:55:51.236013121 +0000
+++ mpfr-3.1.6-b/VERSION	2017-12-15 12:37:44.142256319 +0000
@@ -1 +1 @@
-3.1.6-p1
+3.1.6-p2
diff -Naurd mpfr-3.1.6-a/src/mpfr.h mpfr-3.1.6-b/src/mpfr.h
--- mpfr-3.1.6-a/src/mpfr.h	2017-10-26 13:55:51.232013138 +0000
+++ mpfr-3.1.6-b/src/mpfr.h	2017-12-15 12:37:44.142256319 +0000
@@ -27,7 +27,7 @@
 #define MPFR_VERSION_MAJOR 3
 #define MPFR_VERSION_MINOR 1
 #define MPFR_VERSION_PATCHLEVEL 6
-#define MPFR_VERSION_STRING "3.1.6-p1"
+#define MPFR_VERSION_STRING "3.1.6-p2"
 
 /* Macros dealing with MPFR VERSION */
 #define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.6-a/src/root.c mpfr-3.1.6-b/src/root.c
--- mpfr-3.1.6-a/src/root.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/src/root.c	2017-12-15 12:37:44.118256398 +0000
@@ -107,6 +107,11 @@
       MPFR_RET_NAN;
     }
 
+  /* Special case |x| = 1. Note that if x = -1, then k is odd
+     (NaN results have already been filtered), so that y = -1. */
+  if (mpfr_cmpabs (x, __gmpfr_one) == 0)
+    return mpfr_set (y, x, rnd_mode);
+
   /* General case */
 
   /* For large k, use exp(log(x)/k). The threshold of 100 seems to be quite
@@ -188,6 +193,14 @@
    Assume all special cases have been eliminated before.
    In the extended exponent range, overflows/underflows are not possible.
    Assume x > 0, or x < 0 and k odd.
+   Also assume |x| <> 1 because log(1) = 0, which does not have an exponent
+   and would yield a failure in the error bound computation. A priori, this
+   constraint is quite artificial because if |x| is close enough to 1, then
+   the exponent of log|x| does not need to be used (in the code, err would
+   be 1 in such a domain). So this constraint |x| <> 1 could be avoided in
+   the code. However, this is an exact case easy to detect, so that such a
+   change would be useless. Values very close to 1 are not an issue, since
+   an underflow is not possible before the MPFR_GET_EXP.
 */
 static int
 mpfr_root_aux (mpfr_ptr y, mpfr_srcptr x, unsigned long k, mpfr_rnd_t rnd_mode)
@@ -219,7 +232,8 @@
       mpfr_log (t, absx, MPFR_RNDN);
       /* t = log|x| * (1 + theta) with |theta| <= 2^(-w) */
       mpfr_div_ui (t, t, k, MPFR_RNDN);
-      expt = MPFR_GET_EXP (t);
+      /* No possible underflow in mpfr_log and mpfr_div_ui. */
+      expt = MPFR_GET_EXP (t);  /* assumes t <> 0 */
       /* t = log|x|/k * (1 + theta) + eps with |theta| <= 2^(-w)
          and |eps| <= 1/2 ulp(t), thus the total error is bounded
          by 1.5 * 2^(expt - w) */
diff -Naurd mpfr-3.1.6-a/src/version.c mpfr-3.1.6-b/src/version.c
--- mpfr-3.1.6-a/src/version.c	2017-10-26 13:55:51.232013138 +0000
+++ mpfr-3.1.6-b/src/version.c	2017-12-15 12:37:44.142256319 +0000
@@ -25,5 +25,5 @@
 const char *
 mpfr_get_version (void)
 {
-  return "3.1.6-p1";
+  return "3.1.6-p2";
 }
diff -Naurd mpfr-3.1.6-a/tests/troot.c mpfr-3.1.6-b/tests/troot.c
--- mpfr-3.1.6-a/tests/troot.c	2017-01-01 01:39:09.000000000 +0000
+++ mpfr-3.1.6-b/tests/troot.c	2017-12-15 12:37:44.118256398 +0000
@@ -405,6 +405,26 @@
   mpfr_clears (x, y1, y2, (mpfr_ptr) 0);
 }
 
+static void
+bug20171214 (void)
+{
+  mpfr_t x, y;
+  int inex;
+
+  mpfr_init2 (x, 805);
+  mpfr_init2 (y, 837);
+  mpfr_set_ui (x, 1, MPFR_RNDN);
+  inex = mpfr_root (y, x, 120, MPFR_RNDN);
+  MPFR_ASSERTN (inex == 0);
+  MPFR_ASSERTN (mpfr_cmp_ui (y, 1) == 0);
+  mpfr_set_si (x, -1, MPFR_RNDN);
+  inex = mpfr_root (y, x, 121, MPFR_RNDN);
+  MPFR_ASSERTN (inex == 0);
+  MPFR_ASSERTN (mpfr_cmp_si (y, -1) == 0);
+  mpfr_clear (x);
+  mpfr_clear (y);
+}
+
 int
 main (void)
 {
@@ -415,6 +435,7 @@
 
   tests_start_mpfr ();
 
+  bug20171214 ();
   exact_powers (3, 1000);
   special ();
   bigint ();