rpms/libmpfr
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
88c8d7a699
libmpfr/libmpfr-3.1.3.p5-patchset.patch

1831 lines
85 KiB
Diff
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

diff -Naurd mpfr-3.1.3-a/PATCHES mpfr-3.1.3-b/PATCHES
--- mpfr-3.1.3-a/PATCHES 2015-07-02 10:49:23.950112879 +0000
+++ mpfr-3.1.3-b/PATCHES 2015-07-02 10:49:24.042113845 +0000
@@ -0,0 +1 @@
+lngamma-and-doc
diff -Naurd mpfr-3.1.3-a/VERSION mpfr-3.1.3-b/VERSION
--- mpfr-3.1.3-a/VERSION 2015-06-19 19:55:09.000000000 +0000
+++ mpfr-3.1.3-b/VERSION 2015-07-02 10:49:24.042113845 +0000
@@ -1 +1 @@
-3.1.3
+3.1.3-p1
diff -Naurd mpfr-3.1.3-a/doc/mpfr.texi mpfr-3.1.3-b/doc/mpfr.texi
--- mpfr-3.1.3-a/doc/mpfr.texi 2015-06-19 19:55:11.000000000 +0000
+++ mpfr-3.1.3-b/doc/mpfr.texi 2015-07-02 10:49:24.018113593 +0000
@@ -810,13 +810,17 @@
When the input point is in the closure of the domain of the mathematical
function and an input argument is +0 (resp.@: @minus{}0), one considers
the limit when the corresponding argument approaches 0 from above
-(resp.@: below). If the limit is not defined (e.g., @code{mpfr_log} on
-@minus{}0), the behavior is specified in the description of the MPFR function.
+(resp.@: below), if possible. If the limit is not defined (e.g.,
+@code{mpfr_sqrt} and @code{mpfr_log} on @minus{}0), the behavior is
+specified in the description of the MPFR function, but must be consistent
+with the rule from the above paragraph (e.g., @code{mpfr_log} on @pom{}0
+gives @minus{}Inf).
When the result is equal to 0, its sign is determined by considering the
limit as if the input point were not in the domain: If one approaches 0
from above (resp.@: below), the result is +0 (resp.@: @minus{}0);
-for example, @code{mpfr_sin} on +0 gives +0.
+for example, @code{mpfr_sin} on @minus{}0 gives @minus{}0 and
+@code{mpfr_acos} on 1 gives +0 (in all rounding modes).
In the other cases, the sign is specified in the description of the MPFR
function; for example @code{mpfr_max} on @minus{}0 and +0 gives +0.
@@ -832,8 +836,8 @@
@c that advantages in practice), like for any bug fix.
Example: @code{mpfr_hypot} on (NaN,0) gives NaN, but @code{mpfr_hypot}
on (NaN,+Inf) gives +Inf (as specified in @ref{Special Functions}),
-since for any finite input @var{x}, @code{mpfr_hypot} on (@var{x},+Inf)
-gives +Inf.
+since for any finite or infinite input @var{x}, @code{mpfr_hypot} on
+(@var{x},+Inf) gives +Inf.
@node Exceptions, Memory Handling, Floating-Point Values on Special Numbers, MPFR Basics
@comment node-name, next, previous, up
@@ -1581,7 +1585,8 @@
@deftypefunx int mpfr_add_z (mpfr_t @var{rop}, mpfr_t @var{op1}, mpz_t @var{op2}, mpfr_rnd_t @var{rnd})
@deftypefunx int mpfr_add_q (mpfr_t @var{rop}, mpfr_t @var{op1}, mpq_t @var{op2}, mpfr_rnd_t @var{rnd})
Set @var{rop} to @math{@var{op1} + @var{op2}} rounded in the direction
-@var{rnd}. For types having no signed zero, it is considered unsigned
+@var{rnd}. The IEEE-754 rules are used, in particular for signed zeros.
+But for types having no signed zeros, 0 is considered unsigned
(i.e., (+0) + 0 = (+0) and (@minus{}0) + 0 = (@minus{}0)).
The @code{mpfr_add_d} function assumes that the radix of the @code{double} type
is a power of 2, with a precision at most that declared by the C implementation
@@ -1599,7 +1604,8 @@
@deftypefunx int mpfr_sub_z (mpfr_t @var{rop}, mpfr_t @var{op1}, mpz_t @var{op2}, mpfr_rnd_t @var{rnd})
@deftypefunx int mpfr_sub_q (mpfr_t @var{rop}, mpfr_t @var{op1}, mpq_t @var{op2}, mpfr_rnd_t @var{rnd})
Set @var{rop} to @math{@var{op1} - @var{op2}} rounded in the direction
-@var{rnd}. For types having no signed zero, it is considered unsigned
+@var{rnd}. The IEEE-754 rules are used, in particular for signed zeros.
+But for types having no signed zeros, 0 is considered unsigned
(i.e., (+0) @minus{} 0 = (+0), (@minus{}0) @minus{} 0 = (@minus{}0),
0 @minus{} (+0) = (@minus{}0) and 0 @minus{} (@minus{}0) = (+0)).
The same restrictions than for @code{mpfr_add_d} apply to @code{mpfr_d_sub}
@@ -1615,7 +1621,7 @@
Set @var{rop} to @math{@var{op1} @GMPtimes{} @var{op2}} rounded in the
direction @var{rnd}.
When a result is zero, its sign is the product of the signs of the operands
-(for types having no signed zero, it is considered positive).
+(for types having no signed zeros, 0 is considered positive).
The same restrictions than for @code{mpfr_add_d} apply to @code{mpfr_mul_d}.
@end deftypefun
@@ -1635,7 +1641,7 @@
@deftypefunx int mpfr_div_q (mpfr_t @var{rop}, mpfr_t @var{op1}, mpq_t @var{op2}, mpfr_rnd_t @var{rnd})
Set @var{rop} to @math{@var{op1}/@var{op2}} rounded in the direction @var{rnd}.
When a result is zero, its sign is the product of the signs of the operands
-(for types having no signed zero, it is considered positive).
+(for types having no signed zeros, 0 is considered positive).
The same restrictions than for @code{mpfr_add_d} apply to @code{mpfr_d_div}
and @code{mpfr_div_d}.
@end deftypefun
@@ -1643,15 +1649,18 @@
@deftypefun int mpfr_sqrt (mpfr_t @var{rop}, mpfr_t @var{op}, mpfr_rnd_t @var{rnd})
@deftypefunx int mpfr_sqrt_ui (mpfr_t @var{rop}, unsigned long int @var{op}, mpfr_rnd_t @var{rnd})
Set @var{rop} to @m{\sqrt{@var{op}}, the square root of @var{op}}
-rounded in the direction @var{rnd} (set @var{rop} to @minus{}0 if @var{op} is
-@minus{}0, to be consistent with the IEEE 754 standard).
+rounded in the direction @var{rnd}. Set @var{rop} to @minus{}0 if
+@var{op} is @minus{}0, to be consistent with the IEEE 754 standard.
Set @var{rop} to NaN if @var{op} is negative.
@end deftypefun
@deftypefun int mpfr_rec_sqrt (mpfr_t @var{rop}, mpfr_t @var{op}, mpfr_rnd_t @var{rnd})
Set @var{rop} to @m{1/\sqrt{@var{op}}, the reciprocal square root of @var{op}}
-rounded in the direction @var{rnd}. Set @var{rop} to +Inf if @var{op} is
-@pom{}0, +0 if @var{op} is +Inf, and NaN if @var{op} is negative.
+rounded in the direction @var{rnd}. Set @var{rop} to +Inf if @var{op} is
+@pom{}0, +0 if @var{op} is +Inf, and NaN if @var{op} is negative. Warning!
+Therefore the result on @minus{}0 is different from the one of the rSqrt
+function recommended by the IEEE 754-2008 standard (Section 9.2.1), which
+is @minus{}Inf instead of +Inf.
@end deftypefun
@deftypefun int mpfr_cbrt (mpfr_t @var{rop}, mpfr_t @var{op}, mpfr_rnd_t @var{rnd})
@@ -1832,7 +1841,9 @@
@m{\log_2 @var{op}, log2(@var{op})} or
@m{\log_{10} @var{op}, log10(@var{op})}, respectively,
rounded in the direction @var{rnd}.
-Set @var{rop} to @minus{}Inf if @var{op} is @minus{}0
+Set @var{rop} to +0 if @var{op} is 1 (in all rounding modes),
+for consistency with the ISO C99 and IEEE 754-2008 standards.
+Set @var{rop} to @minus{}Inf if @var{op} is @pom{}0
(i.e., the sign of the zero has no influence on the result).
@end deftypefun
@@ -2003,8 +2014,11 @@
@deftypefun int mpfr_lngamma (mpfr_t @var{rop}, mpfr_t @var{op}, mpfr_rnd_t @var{rnd})
Set @var{rop} to the value of the logarithm of the Gamma function on @var{op},
rounded in the direction @var{rnd}.
-When @math{@minus{}2@var{k}@minus{}1 @le{} @var{op} @le{} @minus{}2@var{k}},
-@var{k} being a non-negative integer, @var{rop} is set to NaN.
+When @var{op} is 1 or 2, set @var{rop} to +0 (in all rounding modes).
+When @var{op} is an infinity or a nonpositive integer, set @var{rop} to +Inf,
+following the general rules on special values.
+When @math{@minus{}2@var{k}@minus{}1 < @var{op} < @minus{}2@var{k}},
+@var{k} being a nonnegative integer, set @var{rop} to NaN@.
See also @code{mpfr_lgamma}.
@end deftypefun
@@ -2012,10 +2026,11 @@
Set @var{rop} to the value of the logarithm of the absolute value of the
Gamma function on @var{op}, rounded in the direction @var{rnd}. The sign
(1 or @minus{}1) of Gamma(@var{op}) is returned in the object pointed to
-by @var{signp}. When @var{op} is an infinity or a non-positive integer, set
-@var{rop} to +Inf. When @var{op} is NaN, @minus{}Inf or a negative integer,
-*@var{signp} is undefined, and when @var{op} is @pom{}0, *@var{signp} is
-the sign of the zero.
+by @var{signp}.
+When @var{op} is 1 or 2, set @var{rop} to +0 (in all rounding modes).
+When @var{op} is an infinity or a nonpositive integer, set @var{rop} to +Inf.
+When @var{op} is NaN, @minus{}Inf or a negative integer, *@var{signp} is
+undefined, and when @var{op} is @pom{}0, *@var{signp} is the sign of the zero.
@end deftypefun
@deftypefun int mpfr_digamma (mpfr_t @var{rop}, mpfr_t @var{op}, mpfr_rnd_t @var{rnd})
@@ -2064,7 +2079,10 @@
@deftypefunx int mpfr_fms (mpfr_t @var{rop}, mpfr_t @var{op1}, mpfr_t @var{op2}, mpfr_t @var{op3}, mpfr_rnd_t @var{rnd})
Set @var{rop} to @math{(@var{op1} @GMPtimes{} @var{op2}) + @var{op3}}
(resp.@: @math{(@var{op1} @GMPtimes{} @var{op2}) - @var{op3}})
-rounded in the direction @var{rnd}.
+rounded in the direction @var{rnd}. Concerning special values (signed zeros,
+infinities, NaN), these functions behave like a multiplication followed by a
+separate addition or subtraction. That is, the fused operation matters only
+for rounding.
@end deftypefun
@deftypefun int mpfr_agm (mpfr_t @var{rop}, mpfr_t @var{op1}, mpfr_t @var{op2}, mpfr_rnd_t @var{rnd})
@@ -2089,8 +2107,8 @@
i.e., $\sqrt{x^2+y^2}$,
@end tex
rounded in the direction @var{rnd}.
-Special values are handled as described in Section F.9.4.3 of
-the ISO C99 and IEEE 754-2008 standards:
+Special values are handled as described in the ISO C99 (Section F.9.4.3)
+and IEEE 754-2008 (Section 9.2.1) standards:
If @var{x} or @var{y} is an infinity, then +Inf is returned in @var{rop},
even if the other number is NaN.
@end deftypefun
diff -Naurd mpfr-3.1.3-a/doc/mpfr.info mpfr-3.1.3-b/doc/mpfr.info
--- mpfr-3.1.3-a/doc/mpfr.info 2015-06-19 19:55:53.000000000 +0000
+++ mpfr-3.1.3-b/doc/mpfr.info 2015-07-02 10:49:38.718267817 +0000
@@ -1,4 +1,4 @@
-This is mpfr.info, produced by makeinfo version 5.2 from mpfr.texi.
+This is mpfr.info, produced by makeinfo version 6.0 from mpfr.texi.
This manual documents how to install and use the Multiple Precision
Floating-Point Reliable Library, version 3.1.3.
@@ -55,7 +55,7 @@
MPFR Copying Conditions
***********************
-The GNU MPFR library (or MPFR for short) is "free"; this means that
+The GNU MPFR library (or MPFR for short) is “free”; this means that
everyone is free to use it and free to redistribute it on a free basis.
The library is not in the public domain; it is copyrighted and there are
restrictions on its distribution, but these restrictions are designed to
@@ -418,7 +418,7 @@
4.2 Nomenclature and Types
==========================
-A "floating-point number", or "float" for short, is an arbitrary
+A “floating-point number”, or “float” for short, is an arbitrary
precision significand (also called mantissa) with a limited precision
exponent. The C data type for such objects is ‘mpfr_t’ (internally
defined as a one-element array of a structure, and ‘mpfr_ptr’ is the C
@@ -432,7 +432,7 @@
to the other functions supported by MPFR. Unless documented otherwise,
the sign bit of a NaN is unspecified.
-The "precision" is the number of bits used to represent the significand
+The “precision” is the number of bits used to represent the significand
of a floating-point number; the corresponding C data type is
‘mpfr_prec_t’. The precision can be any integer between ‘MPFR_PREC_MIN’
and ‘MPFR_PREC_MAX’. In the current implementation, ‘MPFR_PREC_MIN’ is
@@ -446,7 +446,7 @@
may abort, crash or have undefined behavior (depending on your C
implementation).
-The "rounding mode" specifies the way to round the result of a
+The “rounding mode” specifies the way to round the result of a
floating-point operation, in case the exact result can not be
represented exactly in the destination significand; the corresponding C
data type is ‘mpfr_rnd_t’.
@@ -499,14 +499,14 @@
representable numbers, it is rounded to the one with the least
significant bit set to zero. For example, the number 2.5, which is
represented by (10.1) in binary, is rounded to (10.0)=2 with a precision
-of two bits, and not to (11.0)=3. This rule avoids the "drift"
+of two bits, and not to (11.0)=3. This rule avoids the “drift”
phenomenon mentioned by Knuth in volume 2 of The Art of Computer
Programming (Section 4.2.2).
Most MPFR functions take as first argument the destination variable,
as second and following arguments the input variables, as last argument
a rounding mode, and have a return value of type ‘int’, called the
-"ternary value". The value stored in the destination variable is
+“ternary value”. The value stored in the destination variable is
correctly rounded, i.e., MPFR behaves as if it computed the result with
an infinite precision, then rounded it to the precision of this
variable. The input variables are regarded as exact (in particular,
@@ -572,15 +572,18 @@
When the input point is in the closure of the domain of the
mathematical function and an input argument is +0 (resp. −0), one
considers the limit when the corresponding argument approaches 0 from
-above (resp. below). If the limit is not defined (e.g., ‘mpfr_log’ on
-−0), the behavior is specified in the description of the MPFR function.
+above (resp. below), if possible. If the limit is not defined (e.g.,
+‘mpfr_sqrt’ and ‘mpfr_log’ on −0), the behavior is specified in the
+description of the MPFR function, but must be consistent with the rule
+from the above paragraph (e.g., ‘mpfr_log’ on ±0 gives −Inf).
When the result is equal to 0, its sign is determined by considering
the limit as if the input point were not in the domain: If one
approaches 0 from above (resp. below), the result is +0 (resp. −0); for
-example, ‘mpfr_sin’ on +0 gives +0. In the other cases, the sign is
-specified in the description of the MPFR function; for example
-‘mpfr_max’ on −0 and +0 gives +0.
+example, ‘mpfr_sin’ on −0 gives −0 and ‘mpfr_acos’ on 1 gives +0 (in all
+rounding modes). In the other cases, the sign is specified in the
+description of the MPFR function; for example ‘mpfr_max’ on −0 and +0
+gives +0.
When the input point is not in the closure of the domain of the
function, the result is NaN. Example: ‘mpfr_sqrt’ on −17 gives NaN.
@@ -590,8 +593,8 @@
numbers; such a case is always explicitly specified in *note MPFR
Interface::. Example: ‘mpfr_hypot’ on (NaN,0) gives NaN, but
‘mpfr_hypot’ on (NaN,+Inf) gives +Inf (as specified in *note Special
-Functions::), since for any finite input X, ‘mpfr_hypot’ on (X,+Inf)
-gives +Inf.
+Functions::), since for any finite or infinite input X, ‘mpfr_hypot’ on
+(X,+Inf) gives +Inf.

File: mpfr.info, Node: Exceptions, Next: Memory Handling, Prev: Floating-Point Values on Special Numbers, Up: MPFR Basics
@@ -1253,8 +1256,9 @@
mpfr_rnd_t RND)
-- Function: int mpfr_add_q (mpfr_t ROP, mpfr_t OP1, mpq_t OP2,
mpfr_rnd_t RND)
- Set ROP to OP1 + OP2 rounded in the direction RND. For types
- having no signed zero, it is considered unsigned (i.e., (+0) + 0 =
+ Set ROP to OP1 + OP2 rounded in the direction RND. The IEEE-754
+ rules are used, in particular for signed zeros. But for types
+ having no signed zeros, 0 is considered unsigned (i.e., (+0) + 0 =
(+0) and (−0) + 0 = (−0)). The ‘mpfr_add_d’ function assumes that
the radix of the ‘double’ type is a power of 2, with a precision at
most that declared by the C implementation (macro
@@ -1280,8 +1284,9 @@
mpfr_rnd_t RND)
-- Function: int mpfr_sub_q (mpfr_t ROP, mpfr_t OP1, mpq_t OP2,
mpfr_rnd_t RND)
- Set ROP to OP1 - OP2 rounded in the direction RND. For types
- having no signed zero, it is considered unsigned (i.e., (+0) − 0 =
+ Set ROP to OP1 - OP2 rounded in the direction RND. The IEEE-754
+ rules are used, in particular for signed zeros. But for types
+ having no signed zeros, 0 is considered unsigned (i.e., (+0) − 0 =
(+0), (−0) − 0 = (−0), 0 − (+0) = (−0) and 0 − (−0) = (+0)). The
same restrictions than for ‘mpfr_add_d’ apply to ‘mpfr_d_sub’ and
‘mpfr_sub_d’.
@@ -1300,7 +1305,7 @@
mpfr_rnd_t RND)
Set ROP to OP1 times OP2 rounded in the direction RND. When a
result is zero, its sign is the product of the signs of the
- operands (for types having no signed zero, it is considered
+ operands (for types having no signed zeros, 0 is considered
positive). The same restrictions than for ‘mpfr_add_d’ apply to
‘mpfr_mul_d’.
@@ -1327,21 +1332,24 @@
mpfr_rnd_t RND)
Set ROP to OP1/OP2 rounded in the direction RND. When a result is
zero, its sign is the product of the signs of the operands (for
- types having no signed zero, it is considered positive). The same
+ types having no signed zeros, 0 is considered positive). The same
restrictions than for ‘mpfr_add_d’ apply to ‘mpfr_d_div’ and
‘mpfr_div_d’.
-- Function: int mpfr_sqrt (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
-- Function: int mpfr_sqrt_ui (mpfr_t ROP, unsigned long int OP,
mpfr_rnd_t RND)
- Set ROP to the square root of OP rounded in the direction RND (set
- ROP to −0 if OP is −0, to be consistent with the IEEE 754
- standard). Set ROP to NaN if OP is negative.
+ Set ROP to the square root of OP rounded in the direction RND. Set
+ ROP to −0 if OP is −0, to be consistent with the IEEE 754 standard.
+ Set ROP to NaN if OP is negative.
-- Function: int mpfr_rec_sqrt (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
Set ROP to the reciprocal square root of OP rounded in the
direction RND. Set ROP to +Inf if OP is ±0, +0 if OP is +Inf, and
- NaN if OP is negative.
+ NaN if OP is negative. Warning! Therefore the result on −0 is
+ different from the one of the rSqrt function recommended by the
+ IEEE 754-2008 standard (Section 9.2.1), which is −Inf instead of
+ +Inf.
-- Function: int mpfr_cbrt (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
-- Function: int mpfr_root (mpfr_t ROP, mpfr_t OP, unsigned long int K,
@@ -1515,8 +1523,10 @@
-- Function: int mpfr_log2 (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
-- Function: int mpfr_log10 (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
Set ROP to the natural logarithm of OP, log2(OP) or log10(OP),
- respectively, rounded in the direction RND. Set ROP to −Inf if OP
- is −0 (i.e., the sign of the zero has no influence on the result).
+ respectively, rounded in the direction RND. Set ROP to +0 if OP is
+ 1 (in all rounding modes), for consistency with the ISO C99 and
+ IEEE 754-2008 standards. Set ROP to −Inf if OP is ±0 (i.e., the
+ sign of the zero has no influence on the result).
-- Function: int mpfr_exp (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
-- Function: int mpfr_exp2 (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
@@ -1649,17 +1659,21 @@
-- Function: int mpfr_lngamma (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
Set ROP to the value of the logarithm of the Gamma function on OP,
- rounded in the direction RND. When −2K−1 <= OP <= −2K, K being a
- non-negative integer, ROP is set to NaN. See also ‘mpfr_lgamma’.
+ rounded in the direction RND. When OP is 1 or 2, set ROP to +0 (in
+ all rounding modes). When OP is an infinity or a nonpositive
+ integer, set ROP to +Inf, following the general rules on special
+ values. When −2K−1 < OP < −2K, K being a nonnegative integer, set
+ ROP to NaN. See also ‘mpfr_lgamma’.
-- Function: int mpfr_lgamma (mpfr_t ROP, int *SIGNP, mpfr_t OP,
mpfr_rnd_t RND)
Set ROP to the value of the logarithm of the absolute value of the
Gamma function on OP, rounded in the direction RND. The sign (1 or
−1) of Gamma(OP) is returned in the object pointed to by SIGNP.
- When OP is an infinity or a non-positive integer, set ROP to +Inf.
- When OP is NaN, −Inf or a negative integer, *SIGNP is undefined,
- and when OP is ±0, *SIGNP is the sign of the zero.
+ When OP is 1 or 2, set ROP to +0 (in all rounding modes). When OP
+ is an infinity or a nonpositive integer, set ROP to +Inf. When OP
+ is NaN, −Inf or a negative integer, *SIGNP is undefined, and when
+ OP is ±0, *SIGNP is the sign of the zero.
-- Function: int mpfr_digamma (mpfr_t ROP, mpfr_t OP, mpfr_rnd_t RND)
Set ROP to the value of the Digamma (sometimes also called Psi)
@@ -1703,7 +1717,10 @@
-- Function: int mpfr_fms (mpfr_t ROP, mpfr_t OP1, mpfr_t OP2, mpfr_t
OP3, mpfr_rnd_t RND)
Set ROP to (OP1 times OP2) + OP3 (resp. (OP1 times OP2) - OP3)
- rounded in the direction RND.
+ rounded in the direction RND. Concerning special values (signed
+ zeros, infinities, NaN), these functions behave like a
+ multiplication followed by a separate addition or subtraction.
+ That is, the fused operation matters only for rounding.
-- Function: int mpfr_agm (mpfr_t ROP, mpfr_t OP1, mpfr_t OP2,
mpfr_rnd_t RND)
@@ -1717,9 +1734,10 @@
RND)
Set ROP to the Euclidean norm of X and Y, i.e., the square root of
the sum of the squares of X and Y, rounded in the direction RND.
- Special values are handled as described in Section F.9.4.3 of the
- ISO C99 and IEEE 754-2008 standards: If X or Y is an infinity, then
- +Inf is returned in ROP, even if the other number is NaN.
+ Special values are handled as described in the ISO C99 (Section
+ F.9.4.3) and IEEE 754-2008 (Section 9.2.1) standards: If X or Y is
+ an infinity, then +Inf is returned in ROP, even if the other number
+ is NaN.
-- Function: int mpfr_ai (mpfr_t ROP, mpfr_t X, mpfr_rnd_t RND)
Set ROP to the value of the Airy function Ai on X, rounded in the
@@ -2670,7 +2688,7 @@
5.16 Internals
==============
-A "limb" means the part of a multi-precision number that fits in a
+A “limb” means the part of a multi-precision number that fits in a
single word. Usually a limb contains 32 or 64 bits. The C data type
for a limb is ‘mp_limb_t’.
@@ -3140,7 +3158,7 @@
0. PREAMBLE
The purpose of this License is to make a manual, textbook, or other
- functional and useful document "free" in the sense of freedom: to
+ functional and useful document “free” in the sense of freedom: to
assure everyone the effective freedom to copy and redistribute it,
with or without modifying it, either commercially or
noncommercially. Secondarily, this License preserves for the
@@ -3655,9 +3673,9 @@
* Menu:
* mpfr_abs: Basic Arithmetic Functions.
- (line 160)
-* mpfr_acos: Special Functions. (line 51)
-* mpfr_acosh: Special Functions. (line 115)
+ (line 165)
+* mpfr_acos: Special Functions. (line 53)
+* mpfr_acosh: Special Functions. (line 117)
* mpfr_add: Basic Arithmetic Functions.
(line 6)
* mpfr_add_d: Basic Arithmetic Functions.
@@ -3670,15 +3688,15 @@
(line 8)
* mpfr_add_z: Basic Arithmetic Functions.
(line 14)
-* mpfr_agm: Special Functions. (line 210)
-* mpfr_ai: Special Functions. (line 226)
-* mpfr_asin: Special Functions. (line 52)
-* mpfr_asinh: Special Functions. (line 116)
+* mpfr_agm: Special Functions. (line 219)
+* mpfr_ai: Special Functions. (line 236)
+* mpfr_asin: Special Functions. (line 54)
+* mpfr_asinh: Special Functions. (line 118)
* mpfr_asprintf: Formatted Output Functions.
(line 193)
-* mpfr_atan: Special Functions. (line 53)
-* mpfr_atan2: Special Functions. (line 63)
-* mpfr_atanh: Special Functions. (line 117)
+* mpfr_atan: Special Functions. (line 55)
+* mpfr_atan2: Special Functions. (line 65)
+* mpfr_atanh: Special Functions. (line 119)
* mpfr_buildopt_decimal_p: Miscellaneous Functions.
(line 162)
* mpfr_buildopt_gmpinternals_p: Miscellaneous Functions.
@@ -3690,7 +3708,7 @@
* mpfr_can_round: Rounding Related Functions.
(line 39)
* mpfr_cbrt: Basic Arithmetic Functions.
- (line 108)
+ (line 113)
* mpfr_ceil: Integer Related Functions.
(line 7)
* mpfr_check_range: Exception Related Functions.
@@ -3735,18 +3753,18 @@
(line 27)
* mpfr_cmp_z: Comparison Functions.
(line 11)
-* mpfr_const_catalan: Special Functions. (line 237)
-* mpfr_const_euler: Special Functions. (line 236)
-* mpfr_const_log2: Special Functions. (line 234)
-* mpfr_const_pi: Special Functions. (line 235)
+* mpfr_const_catalan: Special Functions. (line 247)
+* mpfr_const_euler: Special Functions. (line 246)
+* mpfr_const_log2: Special Functions. (line 244)
+* mpfr_const_pi: Special Functions. (line 245)
* mpfr_copysign: Miscellaneous Functions.
(line 109)
-* mpfr_cos: Special Functions. (line 29)
-* mpfr_cosh: Special Functions. (line 95)
-* mpfr_cot: Special Functions. (line 47)
-* mpfr_coth: Special Functions. (line 111)
-* mpfr_csc: Special Functions. (line 46)
-* mpfr_csch: Special Functions. (line 110)
+* mpfr_cos: Special Functions. (line 31)
+* mpfr_cosh: Special Functions. (line 97)
+* mpfr_cot: Special Functions. (line 49)
+* mpfr_coth: Special Functions. (line 113)
+* mpfr_csc: Special Functions. (line 48)
+* mpfr_csch: Special Functions. (line 112)
* mpfr_custom_get_exp: Custom Interface. (line 75)
* mpfr_custom_get_kind: Custom Interface. (line 65)
* mpfr_custom_get_significand: Custom Interface. (line 70)
@@ -3756,47 +3774,47 @@
* mpfr_custom_move: Custom Interface. (line 82)
* MPFR_DECL_INIT: Initialization Functions.
(line 74)
-* mpfr_digamma: Special Functions. (line 166)
+* mpfr_digamma: Special Functions. (line 172)
* mpfr_dim: Basic Arithmetic Functions.
- (line 166)
+ (line 171)
* mpfr_div: Basic Arithmetic Functions.
- (line 72)
+ (line 74)
* mpfr_divby0_p: Exception Related Functions.
(line 134)
* mpfr_div_2exp: Compatibility with MPF.
(line 49)
* mpfr_div_2si: Basic Arithmetic Functions.
- (line 181)
+ (line 186)
* mpfr_div_2ui: Basic Arithmetic Functions.
- (line 179)
+ (line 184)
* mpfr_div_d: Basic Arithmetic Functions.
- (line 84)
+ (line 86)
* mpfr_div_q: Basic Arithmetic Functions.
- (line 88)
+ (line 90)
* mpfr_div_si: Basic Arithmetic Functions.
- (line 80)
+ (line 82)
* mpfr_div_ui: Basic Arithmetic Functions.
- (line 76)
+ (line 78)
* mpfr_div_z: Basic Arithmetic Functions.
- (line 86)
+ (line 88)
* mpfr_d_div: Basic Arithmetic Functions.
- (line 82)
+ (line 84)
* mpfr_d_sub: Basic Arithmetic Functions.
- (line 35)
-* mpfr_eint: Special Functions. (line 133)
+ (line 36)
+* mpfr_eint: Special Functions. (line 135)
* mpfr_eq: Compatibility with MPF.
(line 28)
* mpfr_equal_p: Comparison Functions.
(line 59)
* mpfr_erangeflag_p: Exception Related Functions.
(line 137)
-* mpfr_erf: Special Functions. (line 177)
-* mpfr_erfc: Special Functions. (line 178)
-* mpfr_exp: Special Functions. (line 23)
-* mpfr_exp10: Special Functions. (line 25)
-* mpfr_exp2: Special Functions. (line 24)
-* mpfr_expm1: Special Functions. (line 129)
-* mpfr_fac_ui: Special Functions. (line 121)
+* mpfr_erf: Special Functions. (line 183)
+* mpfr_erfc: Special Functions. (line 184)
+* mpfr_exp: Special Functions. (line 25)
+* mpfr_exp10: Special Functions. (line 27)
+* mpfr_exp2: Special Functions. (line 26)
+* mpfr_expm1: Special Functions. (line 131)
+* mpfr_fac_ui: Special Functions. (line 123)
* mpfr_fits_intmax_p: Conversion Functions.
(line 150)
* mpfr_fits_sint_p: Conversion Functions.
@@ -3815,20 +3833,20 @@
(line 147)
* mpfr_floor: Integer Related Functions.
(line 8)
-* mpfr_fma: Special Functions. (line 203)
+* mpfr_fma: Special Functions. (line 209)
* mpfr_fmod: Integer Related Functions.
(line 92)
-* mpfr_fms: Special Functions. (line 205)
+* mpfr_fms: Special Functions. (line 211)
* mpfr_fprintf: Formatted Output Functions.
(line 157)
* mpfr_frac: Integer Related Functions.
(line 76)
-* mpfr_free_cache: Special Functions. (line 244)
+* mpfr_free_cache: Special Functions. (line 254)
* mpfr_free_str: Conversion Functions.
(line 137)
* mpfr_frexp: Conversion Functions.
(line 45)
-* mpfr_gamma: Special Functions. (line 148)
+* mpfr_gamma: Special Functions. (line 150)
* mpfr_get_d: Conversion Functions.
(line 7)
* mpfr_get_decimal64: Conversion Functions.
@@ -3887,7 +3905,7 @@
(line 56)
* mpfr_greater_p: Comparison Functions.
(line 55)
-* mpfr_hypot: Special Functions. (line 218)
+* mpfr_hypot: Special Functions. (line 227)
* mpfr_inexflag_p: Exception Related Functions.
(line 136)
* mpfr_inf_p: Comparison Functions.
@@ -3922,21 +3940,21 @@
(line 31)
* mpfr_integer_p: Integer Related Functions.
(line 119)
-* mpfr_j0: Special Functions. (line 182)
-* mpfr_j1: Special Functions. (line 183)
-* mpfr_jn: Special Functions. (line 184)
+* mpfr_j0: Special Functions. (line 188)
+* mpfr_j1: Special Functions. (line 189)
+* mpfr_jn: Special Functions. (line 190)
* mpfr_lessequal_p: Comparison Functions.
(line 58)
* mpfr_lessgreater_p: Comparison Functions.
(line 64)
* mpfr_less_p: Comparison Functions.
(line 57)
-* mpfr_lgamma: Special Functions. (line 157)
-* mpfr_li2: Special Functions. (line 143)
-* mpfr_lngamma: Special Functions. (line 152)
+* mpfr_lgamma: Special Functions. (line 162)
+* mpfr_li2: Special Functions. (line 145)
+* mpfr_lngamma: Special Functions. (line 154)
* mpfr_log: Special Functions. (line 16)
* mpfr_log10: Special Functions. (line 18)
-* mpfr_log1p: Special Functions. (line 125)
+* mpfr_log1p: Special Functions. (line 127)
* mpfr_log2: Special Functions. (line 17)
* mpfr_max: Miscellaneous Functions.
(line 22)
@@ -3947,29 +3965,29 @@
* mpfr_modf: Integer Related Functions.
(line 82)
* mpfr_mul: Basic Arithmetic Functions.
- (line 51)
+ (line 53)
* mpfr_mul_2exp: Compatibility with MPF.
(line 47)
* mpfr_mul_2si: Basic Arithmetic Functions.
- (line 174)
+ (line 179)
* mpfr_mul_2ui: Basic Arithmetic Functions.
- (line 172)
+ (line 177)
* mpfr_mul_d: Basic Arithmetic Functions.
- (line 57)
+ (line 59)
* mpfr_mul_q: Basic Arithmetic Functions.
- (line 61)
+ (line 63)
* mpfr_mul_si: Basic Arithmetic Functions.
- (line 55)
+ (line 57)
* mpfr_mul_ui: Basic Arithmetic Functions.
- (line 53)
+ (line 55)
* mpfr_mul_z: Basic Arithmetic Functions.
- (line 59)
+ (line 61)
* mpfr_nanflag_p: Exception Related Functions.
(line 135)
* mpfr_nan_p: Comparison Functions.
(line 39)
* mpfr_neg: Basic Arithmetic Functions.
- (line 159)
+ (line 164)
* mpfr_nextabove: Miscellaneous Functions.
(line 15)
* mpfr_nextbelow: Miscellaneous Functions.
@@ -3983,13 +4001,13 @@
* mpfr_overflow_p: Exception Related Functions.
(line 133)
* mpfr_pow: Basic Arithmetic Functions.
- (line 116)
+ (line 121)
* mpfr_pow_si: Basic Arithmetic Functions.
- (line 120)
+ (line 125)
* mpfr_pow_ui: Basic Arithmetic Functions.
- (line 118)
+ (line 123)
* mpfr_pow_z: Basic Arithmetic Functions.
- (line 122)
+ (line 127)
* mpfr_prec_round: Rounding Related Functions.
(line 13)
* ‘mpfr_prec_t’: Nomenclature and Types.
@@ -3999,7 +4017,7 @@
* mpfr_print_rnd_mode: Rounding Related Functions.
(line 71)
* mpfr_rec_sqrt: Basic Arithmetic Functions.
- (line 103)
+ (line 105)
* mpfr_regular_p: Comparison Functions.
(line 43)
* mpfr_reldiff: Compatibility with MPF.
@@ -4021,11 +4039,11 @@
* ‘mpfr_rnd_t’: Nomenclature and Types.
(line 34)
* mpfr_root: Basic Arithmetic Functions.
- (line 109)
+ (line 114)
* mpfr_round: Integer Related Functions.
(line 9)
-* mpfr_sec: Special Functions. (line 45)
-* mpfr_sech: Special Functions. (line 109)
+* mpfr_sec: Special Functions. (line 47)
+* mpfr_sech: Special Functions. (line 111)
* mpfr_set: Assignment Functions.
(line 9)
* mpfr_setsign: Miscellaneous Functions.
@@ -4100,57 +4118,57 @@
(line 49)
* mpfr_signbit: Miscellaneous Functions.
(line 99)
-* mpfr_sin: Special Functions. (line 30)
-* mpfr_sinh: Special Functions. (line 96)
-* mpfr_sinh_cosh: Special Functions. (line 101)
-* mpfr_sin_cos: Special Functions. (line 35)
+* mpfr_sin: Special Functions. (line 32)
+* mpfr_sinh: Special Functions. (line 98)
+* mpfr_sinh_cosh: Special Functions. (line 103)
+* mpfr_sin_cos: Special Functions. (line 37)
* mpfr_si_div: Basic Arithmetic Functions.
- (line 78)
+ (line 80)
* mpfr_si_sub: Basic Arithmetic Functions.
- (line 31)
+ (line 32)
* mpfr_snprintf: Formatted Output Functions.
(line 180)
* mpfr_sprintf: Formatted Output Functions.
(line 170)
* mpfr_sqr: Basic Arithmetic Functions.
- (line 69)
+ (line 71)
* mpfr_sqrt: Basic Arithmetic Functions.
- (line 96)
+ (line 98)
* mpfr_sqrt_ui: Basic Arithmetic Functions.
- (line 97)
+ (line 99)
* mpfr_strtofr: Assignment Functions.
(line 80)
* mpfr_sub: Basic Arithmetic Functions.
- (line 25)
+ (line 26)
* mpfr_subnormalize: Exception Related Functions.
(line 60)
* mpfr_sub_d: Basic Arithmetic Functions.
- (line 37)
+ (line 38)
* mpfr_sub_q: Basic Arithmetic Functions.
- (line 43)
+ (line 44)
* mpfr_sub_si: Basic Arithmetic Functions.
- (line 33)
+ (line 34)
* mpfr_sub_ui: Basic Arithmetic Functions.
- (line 29)
+ (line 30)
* mpfr_sub_z: Basic Arithmetic Functions.
- (line 41)
-* mpfr_sum: Special Functions. (line 252)
+ (line 42)
+* mpfr_sum: Special Functions. (line 262)
* mpfr_swap: Assignment Functions.
(line 150)
* ‘mpfr_t’: Nomenclature and Types.
(line 6)
-* mpfr_tan: Special Functions. (line 31)
-* mpfr_tanh: Special Functions. (line 97)
+* mpfr_tan: Special Functions. (line 33)
+* mpfr_tanh: Special Functions. (line 99)
* mpfr_trunc: Integer Related Functions.
(line 10)
* mpfr_ui_div: Basic Arithmetic Functions.
- (line 74)
+ (line 76)
* mpfr_ui_pow: Basic Arithmetic Functions.
- (line 126)
+ (line 131)
* mpfr_ui_pow_ui: Basic Arithmetic Functions.
- (line 124)
+ (line 129)
* mpfr_ui_sub: Basic Arithmetic Functions.
- (line 27)
+ (line 28)
* mpfr_underflow_p: Exception Related Functions.
(line 132)
* mpfr_unordered_p: Comparison Functions.
@@ -4181,61 +4199,61 @@
(line 182)
* mpfr_vsprintf: Formatted Output Functions.
(line 171)
-* mpfr_y0: Special Functions. (line 193)
-* mpfr_y1: Special Functions. (line 194)
-* mpfr_yn: Special Functions. (line 195)
+* mpfr_y0: Special Functions. (line 199)
+* mpfr_y1: Special Functions. (line 200)
+* mpfr_yn: Special Functions. (line 201)
* mpfr_zero_p: Comparison Functions.
(line 42)
-* mpfr_zeta: Special Functions. (line 171)
-* mpfr_zeta_ui: Special Functions. (line 172)
+* mpfr_zeta: Special Functions. (line 177)
+* mpfr_zeta_ui: Special Functions. (line 178)
* mpfr_z_sub: Basic Arithmetic Functions.
- (line 39)
+ (line 40)

Tag Table:
Node: Top775
Node: Copying2007
-Node: Introduction to MPFR3766
-Node: Installing MPFR5880
-Node: Reporting Bugs11323
-Node: MPFR Basics13353
-Node: Headers and Libraries13669
-Node: Nomenclature and Types16828
-Node: MPFR Variable Conventions18874
-Node: Rounding Modes20418
-Ref: ternary value21544
-Node: Floating-Point Values on Special Numbers23526
-Node: Exceptions26572
-Node: Memory Handling29749
-Node: MPFR Interface30894
-Node: Initialization Functions33008
-Node: Assignment Functions40318
-Node: Combined Initialization and Assignment Functions49673
-Node: Conversion Functions50974
-Node: Basic Arithmetic Functions60035
-Node: Comparison Functions69200
-Node: Special Functions72687
-Node: Input and Output Functions86672
-Node: Formatted Output Functions88644
-Node: Integer Related Functions98431
-Node: Rounding Related Functions105051
-Node: Miscellaneous Functions108888
-Node: Exception Related Functions117568
-Node: Compatibility with MPF124386
-Node: Custom Interface127127
-Node: Internals131526
-Node: API Compatibility133066
-Node: Type and Macro Changes134995
-Node: Added Functions137844
-Node: Changed Functions141132
-Node: Removed Functions145545
-Node: Other Changes145973
-Node: Contributors147576
-Node: References150219
-Node: GNU Free Documentation License151973
-Node: Concept Index174562
-Node: Function and Type Index180659
+Node: Introduction to MPFR3770
+Node: Installing MPFR5884
+Node: Reporting Bugs11327
+Node: MPFR Basics13357
+Node: Headers and Libraries13673
+Node: Nomenclature and Types16832
+Node: MPFR Variable Conventions18894
+Node: Rounding Modes20438
+Ref: ternary value21568
+Node: Floating-Point Values on Special Numbers23554
+Node: Exceptions26813
+Node: Memory Handling29990
+Node: MPFR Interface31135
+Node: Initialization Functions33249
+Node: Assignment Functions40559
+Node: Combined Initialization and Assignment Functions49914
+Node: Conversion Functions51215
+Node: Basic Arithmetic Functions60276
+Node: Comparison Functions69777
+Node: Special Functions73264
+Node: Input and Output Functions87862
+Node: Formatted Output Functions89834
+Node: Integer Related Functions99621
+Node: Rounding Related Functions106241
+Node: Miscellaneous Functions110078
+Node: Exception Related Functions118758
+Node: Compatibility with MPF125576
+Node: Custom Interface128317
+Node: Internals132716
+Node: API Compatibility134260
+Node: Type and Macro Changes136189
+Node: Added Functions139038
+Node: Changed Functions142326
+Node: Removed Functions146739
+Node: Other Changes147167
+Node: Contributors148770
+Node: References151413
+Node: GNU Free Documentation License153167
+Node: Concept Index175760
+Node: Function and Type Index181857

End Tag Table
diff -Naurd mpfr-3.1.3-a/src/lngamma.c mpfr-3.1.3-b/src/lngamma.c
--- mpfr-3.1.3-a/src/lngamma.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/lngamma.c 2015-07-02 10:49:24.018113593 +0000
@@ -603,16 +603,17 @@
mpfr_get_prec (y), mpfr_log_prec, y, inex));
/* special cases */
- if (MPFR_UNLIKELY (MPFR_IS_SINGULAR (x)))
+ if (MPFR_UNLIKELY (MPFR_IS_SINGULAR (x) ||
+ (MPFR_IS_NEG (x) && mpfr_integer_p (x))))
{
- if (MPFR_IS_NAN (x) || MPFR_IS_NEG (x))
+ if (MPFR_IS_NAN (x))
{
MPFR_SET_NAN (y);
MPFR_RET_NAN;
}
- else /* lngamma(+Inf) = lngamma(+0) = +Inf */
+ else /* lngamma(+/-Inf) = lngamma(nonpositive integer) = +Inf */
{
- if (MPFR_IS_ZERO (x))
+ if (!MPFR_IS_INF (x))
mpfr_set_divby0 ();
MPFR_SET_INF (y);
MPFR_SET_POS (y);
@@ -620,8 +621,8 @@
}
}
- /* if x < 0 and -2k-1 <= x <= -2k, then lngamma(x) = NaN */
- if (MPFR_IS_NEG (x) && (unit_bit (x) == 0 || mpfr_integer_p (x)))
+ /* if -2k-1 < x < -2k <= 0, then lngamma(x) = NaN */
+ if (MPFR_IS_NEG (x) && unit_bit (x) == 0)
{
MPFR_SET_NAN (y);
MPFR_RET_NAN;
diff -Naurd mpfr-3.1.3-a/src/mpfr.h mpfr-3.1.3-b/src/mpfr.h
--- mpfr-3.1.3-a/src/mpfr.h 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/mpfr.h 2015-07-02 10:49:24.038113803 +0000
@@ -27,7 +27,7 @@
#define MPFR_VERSION_MAJOR 3
#define MPFR_VERSION_MINOR 1
#define MPFR_VERSION_PATCHLEVEL 3
-#define MPFR_VERSION_STRING "3.1.3"
+#define MPFR_VERSION_STRING "3.1.3-p1"
/* Macros dealing with MPFR VERSION */
#define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.3-a/src/version.c mpfr-3.1.3-b/src/version.c
--- mpfr-3.1.3-a/src/version.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/version.c 2015-07-02 10:49:24.042113845 +0000
@@ -25,5 +25,5 @@
const char *
mpfr_get_version (void)
{
- return "3.1.3";
+ return "3.1.3-p1";
}
diff -Naurd mpfr-3.1.3-a/tests/tlngamma.c mpfr-3.1.3-b/tests/tlngamma.c
--- mpfr-3.1.3-a/tests/tlngamma.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/tests/tlngamma.c 2015-07-02 10:49:24.018113593 +0000
@@ -33,7 +33,7 @@
special (void)
{
mpfr_t x, y;
- int inex;
+ int i, inex;
mpfr_init (x);
mpfr_init (y);
@@ -46,25 +46,29 @@
exit (1);
}
- mpfr_set_inf (x, -1);
+ mpfr_set_inf (x, 1);
+ mpfr_clear_flags ();
mpfr_lngamma (y, x, MPFR_RNDN);
- if (!mpfr_nan_p (y))
+ if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0 || __gmpfr_flags != 0)
{
- printf ("Error for lngamma(-Inf)\n");
+ printf ("Error for lngamma(+Inf)\n");
exit (1);
}
- mpfr_set_inf (x, 1);
+ mpfr_set_inf (x, -1);
+ mpfr_clear_flags ();
mpfr_lngamma (y, x, MPFR_RNDN);
- if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0)
+ if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0 || __gmpfr_flags != 0)
{
- printf ("Error for lngamma(+Inf)\n");
+ printf ("Error for lngamma(-Inf)\n");
exit (1);
}
mpfr_set_ui (x, 0, MPFR_RNDN);
+ mpfr_clear_flags ();
mpfr_lngamma (y, x, MPFR_RNDN);
- if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0)
+ if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0 ||
+ __gmpfr_flags != MPFR_FLAGS_DIVBY0)
{
printf ("Error for lngamma(+0)\n");
exit (1);
@@ -72,32 +76,58 @@
mpfr_set_ui (x, 0, MPFR_RNDN);
mpfr_neg (x, x, MPFR_RNDN);
+ mpfr_clear_flags ();
mpfr_lngamma (y, x, MPFR_RNDN);
- if (!mpfr_nan_p (y))
+ if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0 ||
+ __gmpfr_flags != MPFR_FLAGS_DIVBY0)
{
printf ("Error for lngamma(-0)\n");
exit (1);
}
mpfr_set_ui (x, 1, MPFR_RNDN);
+ mpfr_clear_flags ();
mpfr_lngamma (y, x, MPFR_RNDN);
- if (MPFR_IS_NAN (y) || mpfr_cmp_ui (y, 0) || MPFR_IS_NEG (y))
+ if (mpfr_cmp_ui0 (y, 0) || MPFR_IS_NEG (y))
{
printf ("Error for lngamma(1)\n");
exit (1);
}
- mpfr_set_si (x, -1, MPFR_RNDN);
- mpfr_lngamma (y, x, MPFR_RNDN);
- if (!mpfr_nan_p (y))
+ for (i = 1; i <= 5; i++)
{
- printf ("Error for lngamma(-1)\n");
- exit (1);
+ int c;
+
+ mpfr_set_si (x, -i, MPFR_RNDN);
+ mpfr_clear_flags ();
+ mpfr_lngamma (y, x, MPFR_RNDN);
+ if (!mpfr_inf_p (y) || mpfr_sgn (y) < 0 ||
+ __gmpfr_flags != MPFR_FLAGS_DIVBY0)
+ {
+ printf ("Error for lngamma(-%d)\n", i);
+ exit (1);
+ }
+ if (i & 1)
+ {
+ mpfr_nextabove (x);
+ c = '+';
+ }
+ else
+ {
+ mpfr_nextbelow (x);
+ c = '-';
+ }
+ mpfr_lngamma (y, x, MPFR_RNDN);
+ if (!mpfr_nan_p (y))
+ {
+ printf ("Error for lngamma(-%d%cepsilon)\n", i, c);
+ exit (1);
+ }
}
mpfr_set_ui (x, 2, MPFR_RNDN);
mpfr_lngamma (y, x, MPFR_RNDN);
- if (MPFR_IS_NAN (y) || mpfr_cmp_ui (y, 0) || MPFR_IS_NEG (y))
+ if (mpfr_cmp_ui0 (y, 0) || MPFR_IS_NEG (y))
{
printf ("Error for lngamma(2)\n");
exit (1);
@@ -127,7 +157,7 @@
mpfr_set_str (x, CHECK_X2, 10, MPFR_RNDN);
mpfr_lngamma (y, x, MPFR_RNDN);
mpfr_set_str (x, CHECK_Y2, 10, MPFR_RNDN);
- if (MPFR_IS_NAN (y) || mpfr_cmp (y, x))
+ if (mpfr_cmp0 (y, x))
{
printf ("mpfr_lngamma("CHECK_X2") is wrong:\n"
"expected ");
@@ -143,7 +173,7 @@
mpfr_lngamma (y, x, MPFR_RNDU);
mpfr_set_prec (x, 175);
mpfr_set_str_binary (x, "0.1010001100011101101011001101110010100001000001000001110011000001101100001111001001000101011011100100010101011110100111110101010100010011010010000101010111001100011000101111E7");
- if (MPFR_IS_NAN (y) || mpfr_cmp (x, y))
+ if (mpfr_cmp0 (x, y))
{
printf ("Error in mpfr_lngamma (1)\n");
exit (1);
@@ -155,7 +185,7 @@
mpfr_lngamma (x, y, MPFR_RNDZ);
mpfr_set_prec (y, 21);
mpfr_set_str_binary (y, "0.111000101000001100101E9");
- if (MPFR_IS_NAN (x) || mpfr_cmp (x, y))
+ if (mpfr_cmp0 (x, y))
{
printf ("Error in mpfr_lngamma (120)\n");
printf ("Expected "); mpfr_print_binary (y); puts ("");
@@ -169,7 +199,7 @@
inex = mpfr_lngamma (y, x, MPFR_RNDN);
mpfr_set_prec (x, 206);
mpfr_set_str_binary (x, "0.10000111011000000011100010101001100110001110000111100011000100100110110010001011011110101001111011110110000001010100111011010000000011100110110101100111000111010011110010000100010111101010001101000110101001E13");
- if (MPFR_IS_NAN (y) || mpfr_cmp (x, y))
+ if (mpfr_cmp0 (x, y))
{
printf ("Error in mpfr_lngamma (768)\n");
exit (1);
@@ -185,7 +215,7 @@
mpfr_set_str_binary (x, "0.1100E-66");
mpfr_lngamma (y, x, MPFR_RNDN);
mpfr_set_str_binary (x, "0.1100E6");
- if (MPFR_IS_NAN (y) || mpfr_cmp (x, y))
+ if (mpfr_cmp0 (x, y))
{
printf ("Error for lngamma(0.1100E-66)\n");
exit (1);
@@ -199,7 +229,7 @@
mpfr_lngamma (y, x, MPFR_RNDN);
mpfr_set_prec (x, 32);
mpfr_set_str_binary (x, "-0.10001000111011111011000010100010E207");
- if (MPFR_IS_NAN (y) || mpfr_cmp (x, y))
+ if (mpfr_cmp0 (x, y))
{
printf ("Error for lngamma(-2^199+0.5)\n");
printf ("Got ");
diff -Naurd mpfr-3.1.3-a/PATCHES mpfr-3.1.3-b/PATCHES
--- mpfr-3.1.3-a/PATCHES 2015-07-02 10:50:08.046573308 +0000
+++ mpfr-3.1.3-b/PATCHES 2015-07-02 10:50:08.126574142 +0000
@@ -0,0 +1 @@
+muldiv-2exp-overflow
diff -Naurd mpfr-3.1.3-a/VERSION mpfr-3.1.3-b/VERSION
--- mpfr-3.1.3-a/VERSION 2015-07-02 10:49:24.042113845 +0000
+++ mpfr-3.1.3-b/VERSION 2015-07-02 10:50:08.126574142 +0000
@@ -1 +1 @@
-3.1.3-p1
+3.1.3-p2
diff -Naurd mpfr-3.1.3-a/src/div_2si.c mpfr-3.1.3-b/src/div_2si.c
--- mpfr-3.1.3-a/src/div_2si.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/div_2si.c 2015-07-02 10:50:08.106573933 +0000
@@ -49,7 +49,7 @@
rnd_mode = MPFR_RNDZ;
return mpfr_underflow (y, rnd_mode, MPFR_SIGN(y));
}
- else if (MPFR_UNLIKELY(n < 0 && (__gmpfr_emax < MPFR_EMIN_MIN - n ||
+ else if (MPFR_UNLIKELY(n <= 0 && (__gmpfr_emax < MPFR_EMIN_MIN - n ||
exp > __gmpfr_emax + n)) )
return mpfr_overflow (y, rnd_mode, MPFR_SIGN(y));
diff -Naurd mpfr-3.1.3-a/src/div_2ui.c mpfr-3.1.3-b/src/div_2ui.c
--- mpfr-3.1.3-a/src/div_2ui.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/div_2ui.c 2015-07-02 10:50:08.106573933 +0000
@@ -32,7 +32,7 @@
rnd_mode),
("y[%Pu]=%.*Rg inexact=%d", mpfr_get_prec(y), mpfr_log_prec, y, inexact));
- if (MPFR_UNLIKELY (MPFR_IS_SINGULAR (x)))
+ if (MPFR_UNLIKELY (n == 0 || MPFR_IS_SINGULAR (x)))
return mpfr_set (y, x, rnd_mode);
else
{
diff -Naurd mpfr-3.1.3-a/src/mpfr.h mpfr-3.1.3-b/src/mpfr.h
--- mpfr-3.1.3-a/src/mpfr.h 2015-07-02 10:49:24.038113803 +0000
+++ mpfr-3.1.3-b/src/mpfr.h 2015-07-02 10:50:08.126574142 +0000
@@ -27,7 +27,7 @@
#define MPFR_VERSION_MAJOR 3
#define MPFR_VERSION_MINOR 1
#define MPFR_VERSION_PATCHLEVEL 3
-#define MPFR_VERSION_STRING "3.1.3-p1"
+#define MPFR_VERSION_STRING "3.1.3-p2"
/* Macros dealing with MPFR VERSION */
#define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.3-a/src/mul_2si.c mpfr-3.1.3-b/src/mul_2si.c
--- mpfr-3.1.3-a/src/mul_2si.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/mul_2si.c 2015-07-02 10:50:08.106573933 +0000
@@ -39,7 +39,7 @@
{
mpfr_exp_t exp = MPFR_GET_EXP (x);
MPFR_SETRAW (inexact, y, x, exp, rnd_mode);
- if (MPFR_UNLIKELY( n > 0 && (__gmpfr_emax < MPFR_EMIN_MIN + n ||
+ if (MPFR_UNLIKELY(n >= 0 && (__gmpfr_emax < MPFR_EMIN_MIN + n ||
exp > __gmpfr_emax - n)))
return mpfr_overflow (y, rnd_mode, MPFR_SIGN(y));
else if (MPFR_UNLIKELY(n < 0 && (__gmpfr_emin > MPFR_EMAX_MAX + n ||
diff -Naurd mpfr-3.1.3-a/src/version.c mpfr-3.1.3-b/src/version.c
--- mpfr-3.1.3-a/src/version.c 2015-07-02 10:49:24.042113845 +0000
+++ mpfr-3.1.3-b/src/version.c 2015-07-02 10:50:08.126574142 +0000
@@ -25,5 +25,5 @@
const char *
mpfr_get_version (void)
{
- return "3.1.3-p1";
+ return "3.1.3-p2";
}
diff -Naurd mpfr-3.1.3-a/tests/tmul_2exp.c mpfr-3.1.3-b/tests/tmul_2exp.c
--- mpfr-3.1.3-a/tests/tmul_2exp.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/tests/tmul_2exp.c 2015-07-02 10:50:08.106573933 +0000
@@ -242,6 +242,76 @@
large (MPFR_EMAX_MAX);
}
+/* Cases where the function overflows on n = 0 when rounding is like
+ away from zero. */
+static void
+overflow0 (mpfr_exp_t emax)
+{
+ mpfr_exp_t old_emax;
+ mpfr_t x, y1, y2;
+ int neg, r, op;
+ static char *sop[4] = { "mul_2ui", "mul_2si", "div_2ui", "div_2si" };
+
+ old_emax = mpfr_get_emax ();
+ set_emax (emax);
+
+ mpfr_init2 (x, 8);
+ mpfr_inits2 (6, y1, y2, (mpfr_ptr) 0);
+
+ mpfr_set_inf (x, 1);
+ mpfr_nextbelow (x);
+
+ for (neg = 0; neg <= 1; neg++)
+ {
+ RND_LOOP (r)
+ {
+ int inex1, inex2;
+ unsigned int flags1, flags2;
+
+ /* Even if there isn't an overflow (rounding ~ toward zero),
+ the result is the same as the one of an overflow. */
+ inex1 = mpfr_overflow (y1, (mpfr_rnd_t) r, neg ? -1 : 1);
+ flags1 = MPFR_FLAGS_INEXACT;
+ if (mpfr_inf_p (y1))
+ flags1 |= MPFR_FLAGS_OVERFLOW;
+ for (op = 0; op < 4; op++)
+ {
+ mpfr_clear_flags ();
+ inex2 =
+ op == 0 ? mpfr_mul_2ui (y2, x, 0, (mpfr_rnd_t) r) :
+ op == 1 ? mpfr_mul_2si (y2, x, 0, (mpfr_rnd_t) r) :
+ op == 2 ? mpfr_div_2ui (y2, x, 0, (mpfr_rnd_t) r) :
+ op == 3 ? mpfr_div_2si (y2, x, 0, (mpfr_rnd_t) r) :
+ (MPFR_ASSERTN (0), 0);
+ flags2 = __gmpfr_flags;
+ if (!(mpfr_equal_p (y1, y2) &&
+ SAME_SIGN (inex1, inex2) &&
+ flags1 == flags2))
+ {
+ printf ("Error in overflow0 for %s, mpfr_%s, emax = %"
+ MPFR_EXP_FSPEC "d,\nx = ",
+ mpfr_print_rnd_mode ((mpfr_rnd_t) r), sop[op],
+ (mpfr_eexp_t) emax);
+ mpfr_dump (x);
+ printf ("Expected ");
+ mpfr_dump (y1);
+ printf (" with inex = %d, flags =", inex1);
+ flags_out (flags1);
+ printf ("Got ");
+ mpfr_dump (y2);
+ printf (" with inex = %d, flags =", inex2);
+ flags_out (flags2);
+ exit (1);
+ }
+ }
+ }
+ mpfr_neg (x, x, MPFR_RNDN);
+ }
+
+ mpfr_clears (x, y1, y2, (mpfr_ptr) 0);
+ set_emax (old_emax);
+}
+
int
main (int argc, char *argv[])
{
@@ -334,6 +404,11 @@
underflow0 ();
large0 ();
+ if (mpfr_get_emax () != MPFR_EMAX_MAX)
+ overflow0 (mpfr_get_emax ());
+ overflow0 (MPFR_EMAX_MAX);
+ overflow0 (-1);
+
tests_end_mpfr ();
return 0;
}
diff -Naurd mpfr-3.1.3-a/PATCHES mpfr-3.1.3-b/PATCHES
--- mpfr-3.1.3-a/PATCHES 2015-07-17 08:54:48.592799981 +0000
+++ mpfr-3.1.3-b/PATCHES 2015-07-17 08:54:48.616811495 +0000
@@ -0,0 +1 @@
+muldiv-2exp-underflow
diff -Naurd mpfr-3.1.3-a/VERSION mpfr-3.1.3-b/VERSION
--- mpfr-3.1.3-a/VERSION 2015-07-02 10:50:08.126574142 +0000
+++ mpfr-3.1.3-b/VERSION 2015-07-17 08:54:48.616811495 +0000
@@ -1 +1 @@
-3.1.3-p2
+3.1.3-p3
diff -Naurd mpfr-3.1.3-a/src/div_2si.c mpfr-3.1.3-b/src/div_2si.c
--- mpfr-3.1.3-a/src/div_2si.c 2015-07-02 10:50:08.106573933 +0000
+++ mpfr-3.1.3-b/src/div_2si.c 2015-07-17 08:54:48.608807656 +0000
@@ -45,7 +45,8 @@
if (rnd_mode == MPFR_RNDN &&
(__gmpfr_emin > MPFR_EMAX_MAX - (n - 1) ||
exp < __gmpfr_emin + (n - 1) ||
- (inexact >= 0 && mpfr_powerof2_raw (y))))
+ ((MPFR_IS_NEG (y) ? inexact <= 0 : inexact >= 0) &&
+ mpfr_powerof2_raw (y))))
rnd_mode = MPFR_RNDZ;
return mpfr_underflow (y, rnd_mode, MPFR_SIGN(y));
}
diff -Naurd mpfr-3.1.3-a/src/div_2ui.c mpfr-3.1.3-b/src/div_2ui.c
--- mpfr-3.1.3-a/src/div_2ui.c 2015-07-02 10:50:08.106573933 +0000
+++ mpfr-3.1.3-b/src/div_2ui.c 2015-07-17 08:54:48.608807656 +0000
@@ -44,7 +44,9 @@
if (MPFR_UNLIKELY (n >= diffexp)) /* exp - n <= emin - 1 */
{
if (rnd_mode == MPFR_RNDN &&
- (n > diffexp || (inexact >= 0 && mpfr_powerof2_raw (y))))
+ (n > diffexp ||
+ ((MPFR_IS_NEG (y) ? inexact <= 0 : inexact >= 0) &&
+ mpfr_powerof2_raw (y))))
rnd_mode = MPFR_RNDZ;
return mpfr_underflow (y, rnd_mode, MPFR_SIGN (y));
}
diff -Naurd mpfr-3.1.3-a/src/mpfr.h mpfr-3.1.3-b/src/mpfr.h
--- mpfr-3.1.3-a/src/mpfr.h 2015-07-02 10:50:08.126574142 +0000
+++ mpfr-3.1.3-b/src/mpfr.h 2015-07-17 08:54:48.616811495 +0000
@@ -27,7 +27,7 @@
#define MPFR_VERSION_MAJOR 3
#define MPFR_VERSION_MINOR 1
#define MPFR_VERSION_PATCHLEVEL 3
-#define MPFR_VERSION_STRING "3.1.3-p2"
+#define MPFR_VERSION_STRING "3.1.3-p3"
/* Macros dealing with MPFR VERSION */
#define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.3-a/src/mul_2si.c mpfr-3.1.3-b/src/mul_2si.c
--- mpfr-3.1.3-a/src/mul_2si.c 2015-07-02 10:50:08.106573933 +0000
+++ mpfr-3.1.3-b/src/mul_2si.c 2015-07-17 08:54:48.608807656 +0000
@@ -48,7 +48,8 @@
if (rnd_mode == MPFR_RNDN &&
(__gmpfr_emin > MPFR_EMAX_MAX + (n + 1) ||
exp < __gmpfr_emin - (n + 1) ||
- (inexact >= 0 && mpfr_powerof2_raw (y))))
+ ((MPFR_IS_NEG (y) ? inexact <= 0 : inexact >= 0) &&
+ mpfr_powerof2_raw (y))))
rnd_mode = MPFR_RNDZ;
return mpfr_underflow (y, rnd_mode, MPFR_SIGN(y));
}
diff -Naurd mpfr-3.1.3-a/src/version.c mpfr-3.1.3-b/src/version.c
--- mpfr-3.1.3-a/src/version.c 2015-07-02 10:50:08.126574142 +0000
+++ mpfr-3.1.3-b/src/version.c 2015-07-17 08:54:48.616811495 +0000
@@ -25,5 +25,5 @@
const char *
mpfr_get_version (void)
{
- return "3.1.3-p2";
+ return "3.1.3-p3";
}
diff -Naurd mpfr-3.1.3-a/tests/tmul_2exp.c mpfr-3.1.3-b/tests/tmul_2exp.c
--- mpfr-3.1.3-a/tests/tmul_2exp.c 2015-07-02 10:50:08.106573933 +0000
+++ mpfr-3.1.3-b/tests/tmul_2exp.c 2015-07-17 08:54:48.608807656 +0000
@@ -50,77 +50,82 @@
{
mpfr_t x, y, z1, z2;
mpfr_exp_t emin;
- int i, k;
+ int i, k, s;
int prec;
int rnd;
int div;
int inex1, inex2;
unsigned int flags1, flags2;
- /* Test mul_2si(x, e - k), div_2si(x, k - e) and div_2ui(x, k - e)
- * with emin = e, x = 1 + i/16, i in { -1, 0, 1 }, and k = 1 to 4,
- * by comparing the result with the one of a simple division.
+ /* Test mul_2si(x, e - k), div_2si(x, k - e) and div_2ui(x, k - e) with
+ * emin = e, x = s * (1 + i/16), i in { -1, 0, 1 }, s in { -1, 1 }, and
+ * k = 1 to 4, by comparing the result with the one of a simple division.
*/
emin = mpfr_get_emin ();
set_emin (e);
mpfr_inits2 (8, x, y, (mpfr_ptr) 0);
for (i = 15; i <= 17; i++)
- {
- inex1 = mpfr_set_ui_2exp (x, i, -4, MPFR_RNDN);
- MPFR_ASSERTN (inex1 == 0);
- for (prec = 6; prec >= 3; prec -= 3)
- {
- mpfr_inits2 (prec, z1, z2, (mpfr_ptr) 0);
- RND_LOOP (rnd)
- for (k = 1; k <= 4; k++)
- {
- /* The following one is assumed to be correct. */
- inex1 = mpfr_mul_2si (y, x, e, MPFR_RNDN);
- MPFR_ASSERTN (inex1 == 0);
- inex1 = mpfr_set_ui (z1, 1 << k, MPFR_RNDN);
- MPFR_ASSERTN (inex1 == 0);
- mpfr_clear_flags ();
- /* Do not use mpfr_div_ui to avoid the optimization
- by mpfr_div_2si. */
- inex1 = mpfr_div (z1, y, z1, (mpfr_rnd_t) rnd);
- flags1 = __gmpfr_flags;
-
- for (div = 0; div <= 2; div++)
+ for (s = 1; s >= -1; s -= 2)
+ {
+ inex1 = mpfr_set_si_2exp (x, s * i, -4, MPFR_RNDN);
+ MPFR_ASSERTN (inex1 == 0);
+ for (prec = 6; prec >= 3; prec -= 3)
+ {
+ mpfr_inits2 (prec, z1, z2, (mpfr_ptr) 0);
+ RND_LOOP (rnd)
+ for (k = 1; k <= 4; k++)
{
+ /* The following one is assumed to be correct. */
+ inex1 = mpfr_mul_2si (y, x, e, MPFR_RNDN);
+ MPFR_ASSERTN (inex1 == 0);
+ inex1 = mpfr_set_ui (z1, 1 << k, MPFR_RNDN);
+ MPFR_ASSERTN (inex1 == 0);
mpfr_clear_flags ();
- inex2 = div == 0 ?
- mpfr_mul_2si (z2, x, e - k, (mpfr_rnd_t) rnd) : div == 1 ?
- mpfr_div_2si (z2, x, k - e, (mpfr_rnd_t) rnd) :
- mpfr_div_2ui (z2, x, k - e, (mpfr_rnd_t) rnd);
- flags2 = __gmpfr_flags;
- if (flags1 == flags2 && SAME_SIGN (inex1, inex2) &&
- mpfr_equal_p (z1, z2))
- continue;
- printf ("Error in underflow(");
- if (e == MPFR_EMIN_MIN)
- printf ("MPFR_EMIN_MIN");
- else if (e == emin)
- printf ("default emin");
- else if (e >= LONG_MIN)
- printf ("%ld", (long) e);
- else
- printf ("<LONG_MIN");
- printf (") with mpfr_%s,\nx = %d/16, prec = %d, k = %d, "
- "%s\n", div == 0 ? "mul_2si" : div == 1 ?
- "div_2si" : "div_2ui", i, prec, k,
- mpfr_print_rnd_mode ((mpfr_rnd_t) rnd));
- printf ("Expected ");
- mpfr_out_str (stdout, 16, 0, z1, MPFR_RNDN);
- printf (", inex = %d, flags = %u\n", SIGN (inex1), flags1);
- printf ("Got ");
- mpfr_out_str (stdout, 16, 0, z2, MPFR_RNDN);
- printf (", inex = %d, flags = %u\n", SIGN (inex2), flags2);
- exit (1);
- } /* div */
- } /* k */
- mpfr_clears (z1, z2, (mpfr_ptr) 0);
- } /* prec */
- } /* i */
+ /* Do not use mpfr_div_ui to avoid the optimization
+ by mpfr_div_2si. */
+ inex1 = mpfr_div (z1, y, z1, (mpfr_rnd_t) rnd);
+ flags1 = __gmpfr_flags;
+
+ for (div = 0; div <= 2; div++)
+ {
+ mpfr_clear_flags ();
+ inex2 =
+ div == 0 ?
+ mpfr_mul_2si (z2, x, e - k, (mpfr_rnd_t) rnd) :
+ div == 1 ?
+ mpfr_div_2si (z2, x, k - e, (mpfr_rnd_t) rnd) :
+ mpfr_div_2ui (z2, x, k - e, (mpfr_rnd_t) rnd);
+ flags2 = __gmpfr_flags;
+ if (flags1 == flags2 && SAME_SIGN (inex1, inex2) &&
+ mpfr_equal_p (z1, z2))
+ continue;
+ printf ("Error in underflow(");
+ if (e == MPFR_EMIN_MIN)
+ printf ("MPFR_EMIN_MIN");
+ else if (e == emin)
+ printf ("default emin");
+ else if (e >= LONG_MIN)
+ printf ("%ld", (long) e);
+ else
+ printf ("<LONG_MIN");
+ printf (") with mpfr_%s,\nx = %d/16, prec = %d, k = %d,"
+ " %s\n", div == 0 ? "mul_2si" : div == 1 ?
+ "div_2si" : "div_2ui", s * i, prec, k,
+ mpfr_print_rnd_mode ((mpfr_rnd_t) rnd));
+ printf ("Expected ");
+ mpfr_out_str (stdout, 16, 0, z1, MPFR_RNDN);
+ printf (", inex = %d, flags = %u\n",
+ SIGN (inex1), flags1);
+ printf ("Got ");
+ mpfr_out_str (stdout, 16, 0, z2, MPFR_RNDN);
+ printf (", inex = %d, flags = %u\n",
+ SIGN (inex2), flags2);
+ exit (1);
+ } /* div */
+ } /* k */
+ mpfr_clears (z1, z2, (mpfr_ptr) 0);
+ } /* prec */
+ } /* i */
mpfr_clears (x, y, (mpfr_ptr) 0);
set_emin (emin);
}
diff -Naurd mpfr-3.1.3-a/PATCHES mpfr-3.1.3-b/PATCHES
--- mpfr-3.1.3-a/PATCHES 2015-07-17 08:58:21.094987384 +0000
+++ mpfr-3.1.3-b/PATCHES 2015-07-17 08:58:21.118986898 +0000
@@ -0,0 +1 @@
+frexp
diff -Naurd mpfr-3.1.3-a/VERSION mpfr-3.1.3-b/VERSION
--- mpfr-3.1.3-a/VERSION 2015-07-17 08:54:48.616811495 +0000
+++ mpfr-3.1.3-b/VERSION 2015-07-17 08:58:21.118986898 +0000
@@ -1 +1 @@
-3.1.3-p3
+3.1.3-p4
diff -Naurd mpfr-3.1.3-a/src/frexp.c mpfr-3.1.3-b/src/frexp.c
--- mpfr-3.1.3-a/src/frexp.c 2015-06-19 19:55:09.000000000 +0000
+++ mpfr-3.1.3-b/src/frexp.c 2015-07-17 08:58:21.106987142 +0000
@@ -26,6 +26,13 @@
mpfr_frexp (mpfr_exp_t *exp, mpfr_ptr y, mpfr_srcptr x, mpfr_rnd_t rnd)
{
int inex;
+ unsigned int saved_flags = __gmpfr_flags;
+ MPFR_BLOCK_DECL (flags);
+
+ MPFR_LOG_FUNC
+ (("x[%Pu]=%.*Rg rnd=%d", mpfr_get_prec (x), mpfr_log_prec, x, rnd),
+ ("y[%Pu]=%.*Rg exp=%" MPFR_EXP_FSPEC "d inex=%d", mpfr_get_prec (y),
+ mpfr_log_prec, y, (mpfr_eexp_t) *exp, inex));
if (MPFR_UNLIKELY(MPFR_IS_SINGULAR(x)))
{
@@ -49,8 +56,32 @@
}
}
- inex = mpfr_set (y, x, rnd);
+ MPFR_BLOCK (flags, inex = mpfr_set (y, x, rnd));
+ __gmpfr_flags = saved_flags;
+
+ /* Possible overflow due to the rounding, no possible underflow. */
+
+ if (MPFR_UNLIKELY (MPFR_OVERFLOW (flags)))
+ {
+ int inex2;
+
+ /* An overflow here means that the exponent of y would be larger than
+ the one of x, thus x would be rounded to the next power of 2, and
+ the returned y should be 1/2 in absolute value, rounded (i.e. with
+ possible underflow or overflow). This also implies that x and y are
+ different objects, so that the exponent of x has not been lost. */
+ MPFR_LOG_MSG (("Internal overflow\n", 0));
+ MPFR_ASSERTD (x != y);
+ *exp = MPFR_GET_EXP (x) + 1;
+ inex2 = mpfr_set_si_2exp (y, MPFR_INT_SIGN (x), -1, rnd);
+ MPFR_LOG_MSG (("inex=%d inex2=%d\n", inex, inex2));
+ if (inex2 != 0)
+ inex = inex2;
+ MPFR_RET (inex);
+ }
+
*exp = MPFR_GET_EXP (y);
- MPFR_SET_EXP (y, 0);
+ /* Do not use MPFR_SET_EXP because the range has not been checked yet. */
+ MPFR_EXP (y) = 0;
return mpfr_check_range (y, inex, rnd);
}
diff -Naurd mpfr-3.1.3-a/src/mpfr.h mpfr-3.1.3-b/src/mpfr.h
--- mpfr-3.1.3-a/src/mpfr.h 2015-07-17 08:54:48.616811495 +0000
+++ mpfr-3.1.3-b/src/mpfr.h 2015-07-17 08:58:21.114986979 +0000
@@ -27,7 +27,7 @@
#define MPFR_VERSION_MAJOR 3
#define MPFR_VERSION_MINOR 1
#define MPFR_VERSION_PATCHLEVEL 3
-#define MPFR_VERSION_STRING "3.1.3-p3"
+#define MPFR_VERSION_STRING "3.1.3-p4"
/* Macros dealing with MPFR VERSION */
#define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.3-a/src/version.c mpfr-3.1.3-b/src/version.c
--- mpfr-3.1.3-a/src/version.c 2015-07-17 08:54:48.616811495 +0000
+++ mpfr-3.1.3-b/src/version.c 2015-07-17 08:58:21.118986898 +0000
@@ -25,5 +25,5 @@
const char *
mpfr_get_version (void)
{
- return "3.1.3-p3";
+ return "3.1.3-p4";
}
diff -Naurd mpfr-3.1.3-a/tests/tfrexp.c mpfr-3.1.3-b/tests/tfrexp.c
--- mpfr-3.1.3-a/tests/tfrexp.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/tests/tfrexp.c 2015-07-17 08:58:21.106987142 +0000
@@ -129,12 +129,115 @@
mpfr_clear (x);
}
+static void check1 (void)
+{
+ mpfr_exp_t emin, emax, e;
+ mpfr_t x, y1, y2;
+ int r, neg, red;
+
+ emin = mpfr_get_emin ();
+ emax = mpfr_get_emax ();
+ set_emin (MPFR_EMIN_MIN);
+ set_emax (MPFR_EMAX_MAX);
+
+ mpfr_init2 (x, 7);
+ mpfr_inits2 (4, y1, y2, (mpfr_ptr) 0);
+
+ mpfr_set_ui_2exp (x, 1, -2, MPFR_RNDN);
+ while (mpfr_regular_p (x))
+ {
+ /* Test the exponents up to 3 and with the maximum exponent
+ (to check potential intermediate overflow). */
+ if (MPFR_GET_EXP (x) == 4)
+ mpfr_set_exp (x, MPFR_EMAX_MAX);
+ e = MPFR_GET_EXP (x);
+ for (neg = 0; neg < 2; neg++)
+ {
+ RND_LOOP (r)
+ {
+ int inex1, inex2;
+ mpfr_exp_t e1, e2;
+ unsigned int flags1, flags2;
+
+ for (red = 0; red < 2; red++)
+ {
+ if (red)
+ {
+ /* e1: exponent of the rounded value of x. */
+ MPFR_ASSERTN (e1 == e || e1 == e + 1);
+ set_emin (e);
+ set_emax (e);
+ mpfr_clear_flags ();
+ inex1 = e1 < 0 ?
+ mpfr_mul_2ui (y1, x, -e1, (mpfr_rnd_t) r) :
+ mpfr_div_2ui (y1, x, e1, (mpfr_rnd_t) r);
+ flags1 = __gmpfr_flags;
+ }
+ else
+ {
+ inex1 = mpfr_set (y1, x, (mpfr_rnd_t) r);
+ e1 = MPFR_IS_INF (y1) ? e + 1 : MPFR_GET_EXP (y1);
+ flags1 = inex1 != 0 ? MPFR_FLAGS_INEXACT : 0;
+ }
+ mpfr_clear_flags ();
+ inex2 = mpfr_frexp (&e2, y2, x, (mpfr_rnd_t) r);
+ flags2 = __gmpfr_flags;
+ set_emin (MPFR_EMIN_MIN);
+ set_emax (MPFR_EMAX_MAX);
+ if ((!red || e == 0) &&
+ (! mpfr_regular_p (y2) || MPFR_GET_EXP (y2) != 0))
+ {
+ printf ("Error in check1 for %s, red = %d, x = ",
+ mpfr_print_rnd_mode ((mpfr_rnd_t) r), red);
+ mpfr_dump (x);
+ printf ("Expected 1/2 <= |y| < 1, got y = ");
+ mpfr_dump (y2);
+ exit (1);
+ }
+ if (!red)
+ {
+ if (e2 > 0)
+ mpfr_mul_2ui (y2, y2, e2, MPFR_RNDN);
+ else if (e2 < 0)
+ mpfr_div_2ui (y2, y2, -e2, MPFR_RNDN);
+ }
+ if (! (SAME_SIGN (inex1, inex2) &&
+ mpfr_equal_p (y1, y2) &&
+ flags1 == flags2))
+ {
+ printf ("Error in check1 for %s, red = %d, x = ",
+ mpfr_print_rnd_mode ((mpfr_rnd_t) r), red);
+ mpfr_dump (x);
+ printf ("Expected y1 = ");
+ mpfr_dump (y1);
+ printf ("Got y2 = ");
+ mpfr_dump (y2);
+ printf ("Expected inex ~= %d, got %d\n", inex1, inex2);
+ printf ("Expected flags:");
+ flags_out (flags1);
+ printf ("Got flags: ");
+ flags_out (flags2);
+ exit (1);
+ }
+ }
+ }
+ mpfr_neg (x, x, MPFR_RNDN);
+ }
+ mpfr_nextabove (x);
+ }
+
+ mpfr_clears (x, y1, y2, (mpfr_ptr) 0);
+ set_emin (emin);
+ set_emax (emax);
+}
+
int
main (int argc, char *argv[])
{
tests_start_mpfr ();
check_special ();
+ check1 ();
tests_end_mpfr ();
return 0;
diff -Naurd mpfr-3.1.3-a/PATCHES mpfr-3.1.3-b/PATCHES
--- mpfr-3.1.3-a/PATCHES 2015-10-29 13:47:46.735901185 +0000
+++ mpfr-3.1.3-b/PATCHES 2015-10-29 13:47:46.763900609 +0000
@@ -0,0 +1 @@
+divhigh-basecase
diff -Naurd mpfr-3.1.3-a/VERSION mpfr-3.1.3-b/VERSION
--- mpfr-3.1.3-a/VERSION 2015-07-17 08:58:21.118986898 +0000
+++ mpfr-3.1.3-b/VERSION 2015-10-29 13:47:46.763900609 +0000
@@ -1 +1 @@
-3.1.3-p4
+3.1.3-p5
diff -Naurd mpfr-3.1.3-a/src/mpfr.h mpfr-3.1.3-b/src/mpfr.h
--- mpfr-3.1.3-a/src/mpfr.h 2015-07-17 08:58:21.114986979 +0000
+++ mpfr-3.1.3-b/src/mpfr.h 2015-10-29 13:47:46.759900692 +0000
@@ -27,7 +27,7 @@
#define MPFR_VERSION_MAJOR 3
#define MPFR_VERSION_MINOR 1
#define MPFR_VERSION_PATCHLEVEL 3
-#define MPFR_VERSION_STRING "3.1.3-p4"
+#define MPFR_VERSION_STRING "3.1.3-p5"
/* Macros dealing with MPFR VERSION */
#define MPFR_VERSION_NUM(a,b,c) (((a) << 16L) | ((b) << 8) | (c))
diff -Naurd mpfr-3.1.3-a/src/mulders.c mpfr-3.1.3-b/src/mulders.c
--- mpfr-3.1.3-a/src/mulders.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/src/mulders.c 2015-10-29 13:47:46.751900855 +0000
@@ -236,9 +236,10 @@
that in addition to the limb np[n-1] to reduce, we have at least 2
extra limbs, thus accessing np[n-3] is valid. */
- /* warning: we can have np[n-1]=d1 and np[n-2]=d0, but since {np,n} < D,
- the largest possible partial quotient is B-1 */
- if (MPFR_UNLIKELY(np[n - 1] == d1 && np[n - 2] == d0))
+ /* Warning: we can have np[n-1]>d1 or (np[n-1]=d1 and np[n-2]>=d0) here,
+ since we truncate the divisor at each step, but since {np,n} < D
+ originally, the largest possible partial quotient is B-1. */
+ if (MPFR_UNLIKELY(np[n-1] > d1 || (np[n-1] == d1 && np[n-2] >= d0)))
q2 = ~ (mp_limb_t) 0;
else
udiv_qr_3by2 (q2, q1, q0, np[n - 1], np[n - 2], np[n - 3],
diff -Naurd mpfr-3.1.3-a/src/version.c mpfr-3.1.3-b/src/version.c
--- mpfr-3.1.3-a/src/version.c 2015-07-17 08:58:21.118986898 +0000
+++ mpfr-3.1.3-b/src/version.c 2015-10-29 13:47:46.763900609 +0000
@@ -25,5 +25,5 @@
const char *
mpfr_get_version (void)
{
- return "3.1.3-p4";
+ return "3.1.3-p5";
}
diff -Naurd mpfr-3.1.3-a/tests/tdiv.c mpfr-3.1.3-b/tests/tdiv.c
--- mpfr-3.1.3-a/tests/tdiv.c 2015-06-19 19:55:10.000000000 +0000
+++ mpfr-3.1.3-b/tests/tdiv.c 2015-10-29 13:47:46.751900855 +0000
@@ -1099,6 +1099,69 @@
mpfr_set_emax (old_emax);
}
+/* Bug in mpfr_divhigh_n_basecase when all limbs of q (except the most
+ significant one) are B-1 where B=2^GMP_NUMB_BITS. Since we truncate
+ the divisor at each step, it might happen at some point that
+ (np[n-1],np[n-2]) > (d1,d0), and not only the equality.
+ Reported by Ricky Farr
+ <https://sympa.inria.fr/sympa/arc/mpfr/2015-10/msg00023.html>
+ To get a failure, a MPFR_DIVHIGH_TAB entry below the MPFR_DIV_THRESHOLD
+ limit must have a value 0. With most mparam.h files, this cannot occur. */
+static void
+test_20151023 (void)
+{
+ mpfr_prec_t p;
+ mpfr_t n, d, q, q0;
+ int inex, i;
+
+ for (p = GMP_NUMB_BITS; p <= 2000; p++)
+ {
+ mpfr_init2 (n, 2*p);
+ mpfr_init2 (d, p);
+ mpfr_init2 (q, p);
+ mpfr_init2 (q0, GMP_NUMB_BITS);
+
+ /* generate a random divisor of p bits */
+ mpfr_urandomb (d, RANDS);
+ /* generate a random quotient of GMP_NUMB_BITS bits */
+ mpfr_urandomb (q0, RANDS);
+ /* zero-pad the quotient to p bits */
+ inex = mpfr_prec_round (q0, p, MPFR_RNDN);
+ MPFR_ASSERTN(inex == 0);
+
+ for (i = 0; i < 3; i++)
+ {
+ /* i=0: try with the original quotient xxx000...000
+ i=1: try with the original quotient minus one ulp
+ i=2: try with the original quotient plus one ulp */
+ if (i == 1)
+ mpfr_nextbelow (q0);
+ else if (i == 2)
+ {
+ mpfr_nextabove (q0);
+ mpfr_nextabove (q0);
+ }
+
+ inex = mpfr_mul (n, d, q0, MPFR_RNDN);
+ MPFR_ASSERTN(inex == 0);
+ mpfr_nextabove (n);
+ mpfr_div (q, n, d, MPFR_RNDN);
+ MPFR_ASSERTN(mpfr_cmp (q, q0) == 0);
+
+ inex = mpfr_mul (n, d, q0, MPFR_RNDN);
+ MPFR_ASSERTN(inex == 0);
+ mpfr_nextbelow (n);
+ mpfr_div (q, n, d, MPFR_RNDN);
+ MPFR_ASSERTN(mpfr_cmp (q, q0) == 0);
+ }
+
+ mpfr_clear (n);
+ mpfr_clear (d);
+ mpfr_clear (q);
+ mpfr_clear (q0);
+ }
+}
+
#define TEST_FUNCTION test_div
#define TWO_ARGS
#define RAND_FUNCTION(x) mpfr_random2(x, MPFR_LIMB_SIZE (x), randlimb () % 100, RANDS)
@@ -1219,6 +1282,7 @@
consistency ();
test_20070603 ();
test_20070628 ();
+ test_20151023 ();
test_generic (2, 800, 50);
test_extreme ();
Reference in New Issue View Git Blame Copy Permalink