The ldbl-128ibm expl wrapper checks the argument to determine when to
call __kernel_standard_l, thereby overriding overflowing results from
__ieee754_expl that could otherwise (given appropriately patched
libgcc) be correct for the rounding mode. This patch changes it to
check the result of __ieee754_expl instead, as other versions of this
wrapper do.
Tested for powerpc.
[BZ #19078]
* sysdeps/ieee754/ldbl-128ibm/w_expl.c (o_thres): Remove variable.
(u_thres): Likewise.
(__expl): Determine whether to call __kernel_standard_l based on
value of result, not argument.
The ldbl-128ibm implementation of logl produces a zero with the wrong
sign for logl (1) in FE_DOWNWARD mode. This patch makes it explicitly
return 0.0L in that case, as in e.g. the ldbl-128 implementation.
Tested for powerpc.
[BZ #19077]
* sysdeps/ieee754/ldbl-128ibm/e_logl.c (__ieee754_logl): Return
0.0L for argument 1.0L.
The ldbl-128ibm implementation of log1pl produces an infinity with the
wrong sign for log1pl (-1) in FE_DOWNWARD mode. This patch fixes this
by changing a division (-1.0L / (x - x)) (incorrect in FE_DOWNWARD
mode) to (-1.0L / 0.0L) (correct in all rounding modes).
Tested for powerpc.
[BZ #19076]
* sysdeps/ieee754/ldbl-128ibm/s_log1pl.c (__log1pl): Divide by
constant 0.0L when computing infinite result.
The ldbl-96 version of lroundl is incorrect for systems with 64-bit
long when the argument's absolute value is just below a power of 2,
2^32 or more, and rounds up to the next integer; in such cases, it
returns 0. The problem is incrementing the high part of the mantissa
loses the high bit of the value (which is not an issue for any other
floating-point format, and is handled specially in lround when the bit
corresponding to 0.5 was in the high part rather than the low part).
This patch fixes this in a similar way to that used in llroundl:
storing the high part in an unsigned long variable before incrementing
it, so problems cannot occur in the case when this code is reachable.
I improved test coverage for both lround and llround by making them
use the same test inputs (appropriately conditioned on the size of
long in the lround case) - complete with the same comments, to make
comparison as easy as possible. (This test coverage improvement was
how I found the lroundl bug.)
Tested for x86_64 and x86.
[BZ #19071]
* sysdeps/ieee754/ldbl-96/s_lroundl.c (__lroundl): Use unsigned
long int variable to store possibly incremented high part of
mantissa.
* math/libm-test.inc (lround_test_data): Add tests used for
llround. Use [LONG_MAX > 0x7fffffff] consistently as condition
for tests requiring 64-bit long. Do not condition tests on
[TEST_FLOAT] unnecessarily.
(llround_test_data): Add tests used for lround. Add another
expectation for the "inexact" exception. Do not condition tests
on [TEST_FLOAT] unnecessarily.
On powerpc32 hard-float, older processors (ones where fcfid is not
available for 32-bit code), GCC generates conversions from integers to
floating point that wrongly convert integer 0 to -0 instead of +0 in
FE_DOWNWARD mode. This in turn results in logb and a few other
functions wrongly returning -0 when they should return +0.
This patch works around this issue in glibc as I proposed in
<https://sourceware.org/ml/libc-alpha/2015-09/msg00728.html>, so that
the affected functions can be correct and the affected tests pass in
the absence of a GCC fix for this longstanding issue (GCC bug 67771 -
if fixed, of course we can put in GCC version conditionals, and
eventually phase out the workarounds). A new macro
FIX_INT_FP_CONVERT_ZERO is added in a new sysdeps header
fix-int-fp-convert-zero.h, and the powerpc32/fpu version of that
header defines the macro based on the results of a configure test for
whether such conversions use the fcfid instruction.
Tested for x86_64 (that installed stripped shared libraries are
unchanged by the patch) and powerpc (that HAVE_PPC_FCFID comes out to
0 as expected and that the relevant tests are fixed). Also tested a
build with GCC configured for -mcpu=power4 and verified that
HAVE_PPC_FCFID comes out to 1 in that case.
There are still some other issues to fix to get test-float and
test-double passing cleanly for older powerpc32 processors (apart from
the need for an ulps regeneration for powerpc). (test-ldouble will be
harder to get passing cleanly, but with a combination of selected
fixes to ldbl-128ibm code that don't involve significant performance
issues, allowing spurious underflow and inexact exceptions for that
format, and lots of XFAILing for the default case of unpatched libgcc,
it should be doable.)
[BZ #887]
[BZ #19049]
[BZ #19050]
* sysdeps/generic/fix-int-fp-convert-zero.h: New file.
* sysdeps/ieee754/dbl-64/e_log10.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log10): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/dbl-64/e_log2.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log2): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/dbl-64/s_erf.c: Include
<fix-int-fp-convert-zero.h>.
(__erfc): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/dbl-64/s_logb.c: Include
<fix-int-fp-convert-zero.h>.
(__logb): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/e_log10f.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log10f): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/e_log2f.c: Include
<fix-int-fp-convert-zero.h>.
(__ieee754_log2f): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/s_erff.c: Include
<fix-int-fp-convert-zero.h>.
(__erfcf): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/flt-32/s_logbf.c: Include
<fix-int-fp-convert-zero.h>.
(__logbf): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/ldbl-128ibm/s_erfl.c: Include
<fix-int-fp-convert-zero.h>.
(__erfcl): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/ieee754/ldbl-128ibm/s_logbl.c: Include
<fix-int-fp-convert-zero.h>.
(__logbl): Adjust signs as needed if FIX_INT_FP_CONVERT_ZERO.
* sysdeps/powerpc/powerpc32/fpu/configure.ac: New file.
* sysdeps/powerpc/powerpc32/fpu/configure: New generated file.
* sysdeps/powerpc/powerpc32/fpu/fix-int-fp-convert-zero.h: New
file.
* config.h.in [_LIBC] (HAVE_PPC_FCFID): New macro.
ISO C requires overflowing results from nexttoward to be the
appropriate infinity independent of the rounding mode, but some
implementations use a rounding-mode-dependent result (this is the same
issue as was fixed for nextafter in bug 16677). This patch fixes the
problem by making the nexttoward implementations discard the result
from the floating-point computation that forced an overflow exception
and then return the infinity previously computed with integer
arithmetic.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #19059]
* math/s_nexttowardf.c (__nexttowardf): Do not return value from
overflowing computation.
* sysdeps/i386/fpu/s_nexttoward.c (__nexttoward): Likewise.
* sysdeps/i386/fpu/s_nexttowardf.c (__nexttowardf): Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttoward.c (__nexttoward):
Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttoward.c (__nexttoward):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttoward.c (__nexttoward): Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-opt/s_nexttowardfd.c (__nldbl_nexttowardf):
Likewise.
* math/libm-test.inc (nexttoward_test_data): Add more tests.
The ldbl-128 / ldbl-128ibm implementation of lgamma has problems with
its handling of large arguments. It has an overflow threshold that is
correct only for ldbl-128, despite being used for both types - with
diagnostic control macros as a temporary measure to disable warnings
about that constant overflowing for ldbl-128ibm - and it has a
calculation that's roughly x * log(x) - x, resulting in overflows for
arguments that are roughly at most a factor 1/log(threshold) below the
overflow threshold.
This patch fixes both issues, using an overflow threshold appropriate
for the type in question and adding another case for large arguments
that avoids the possible intermediate overflow.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16347]
[BZ #19046]
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c: Do not include
<libc-internal.h>.
(MAXLGM): Do not use diagnostic control macros.
[LDBL_MANT_DIG == 106] (MAXLGM): Change value to overflow
threshold for ldbl-128ibm.
(__ieee754_lgammal_r): For large arguments, multiply by log - 1
instead of multiplying by log then subtracting.
* math/auto-libm-test-in: Add more tests of lgamma.
* math/auto-libm-test-out: Regenerated.
The ldbl-128ibm implementation of exp10l uses a version of log(10)
split into high and low parts - but the low part is negative, so
causing spurious overflows from __ieee754_expl (exp_high) in cases
close to the overflow threshold (I added relevant tests close to the
overflow threshold to the testsuite earlier today). The same issue
applies close to the underflow threshold as well (except that spurious
underflows in IBM long double arithmetic are harder to fix than the
other deficiencies, so we might end up permitting those for IBM long
double in the libm testsuite, as permitted by ISO C).
This patch fixes it to use a low part rounded downward to 48 bits
instead. (The choice of 48 instead of 53 bits is to make it more
obviously safe even when the low part of the argument is negative.)
Tested for powerpc. (Note that because of libgcc bugs with
multiplication very close to LDBL_MAX, libgcc also needs patching for
all the problem cases to be fixed, but this patch is still safe and
correct in the absence of such libgcc fixes.)
[BZ #16620]
* sysdeps/ieee754/ldbl-128ibm/e_exp10l.c (log10_high): Use value
of log (10) rounded downward to 48 bits.
(log10_low): Use corresponding low part of log (10).
For arguments with X^2 + Y^2 close to 1, clog and clog10 avoid large
errors from log(hypot) by computing X^2 + Y^2 - 1 in a way that avoids
cancellation error and then using log1p.
However, the thresholds for using that approach still result in log
being used on argument as large as sqrt(13/16) > 0.9, leading to
significant errors, in some cases above the 9ulp maximum allowed in
glibc libm. This patch arranges for the approach using log1p to be
used in any cases where |X|, |Y| < 1 and X^2 + Y^2 >= 0.5 (with the
existing allowance for cases where one of X and Y is very small),
adjusting the __x2y2m1 functions to work with the wider range of
inputs. This way, log only gets used on arguments below sqrt(1/2) (or
substantially above 1), where the error involved is much less.
Tested for x86_64, x86, mips64 and powerpc. For the ulps regeneration
I removed the existing clog and clog10 ulps before regenerating to
allow any reduced ulps to appear. Tests added include those found by
random test generation to produce large ulps either before or after
the patch, and some found by trying inputs close to the (0.75, 0.5)
threshold where the potential errors from using log are largest.
[BZ #19016]
* sysdeps/generic/math_private.h (__x2y2m1f): Update comment to
allow more cases with X^2 + Y^2 >= 0.5.
* sysdeps/ieee754/dbl-64/x2y2m1.c (__x2y2m1): Likewise. Add -1 as
normal element in sum instead of special-casing based on values of
arguments.
* sysdeps/ieee754/dbl-64/x2y2m1f.c (__x2y2m1f): Update comment.
* sysdeps/ieee754/ldbl-128/x2y2m1l.c (__x2y2m1l): Likewise. Add
-1 as normal element in sum instead of special-casing based on
values of arguments.
* sysdeps/ieee754/ldbl-128ibm/x2y2m1l.c (__x2y2m1l): Likewise.
* sysdeps/ieee754/ldbl-96/x2y2m1.c [FLT_EVAL_METHOD != 0]
(__x2y2m1): Update comment.
* sysdeps/ieee754/ldbl-96/x2y2m1l.c (__x2y2m1l): Likewise. Add -1
as normal element in sum instead of special-casing based on values
of arguments.
* math/s_clog.c (__clog): Handle more cases using log1p without
hypot.
* math/s_clog10.c (__clog10): Likewise.
* math/s_clog10f.c (__clog10f): Likewise.
* math/s_clog10l.c (__clog10l): Likewise.
* math/s_clogf.c (__clogf): Likewise.
* math/s_clogl.c (__clogl): Likewise.
* math/auto-libm-test-in: Add more tests of clog and clog10.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
The flt-32 version of powf can be inaccurate because of bugs in the
extra-precision calculation of (x-1)/(x+1) or (x-1.5)/(x+1.5) as part
of calculating log(x) with extra precision: a constant used (as part
of adding 1 or 1.5 through integer arithmetic) is incorrect, and then
the code fails to mask a computed high part before using it in
arithmetic that relies on s_h*t_h being exactly representable. This
patch fixes these bugs.
Tested for x86_64 and x86. x86_64 ulps for powf removed and
regenerated to reflect reduced ulps from the increased accuracy for
existing tests.
[BZ #18956]
* sysdeps/ieee754/flt-32/e_powf.c (__ieee754_powf): Add 0x00400000
not 0x0040000 for high bit of mantissa. Mask with 0xfffff000 when
extracting high part.
* math/auto-libm-test-in: Add another test of pow.
* math/auto-libm-test-out: Regenerated.
* sysdeps/x86_64/fpu/libm-test-ulps: Update.
Similar to various other bugs in this area, pow functions can fail to
raise the underflow exception when the result is tiny and inexact but
one or more low bits of the intermediate result that is scaled down
(or, in the i386 case, converted from a wider evaluation format) are
zero. This patch forces the exception in a similar way to previous
fixes, thereby concluding the fixes for known bugs with missing
underflow exceptions currently filed in Bugzilla.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18825]
* sysdeps/i386/fpu/i386-math-asm.h (FLT_NARROW_EVAL_UFLOW_NONNAN):
New macro.
(DBL_NARROW_EVAL_UFLOW_NONNAN): Likewise.
(LDBL_CHECK_FORCE_UFLOW_NONNAN): Likewise.
* sysdeps/i386/fpu/e_pow.S: Use DEFINE_DBL_MIN.
(__ieee754_pow): Use DBL_NARROW_EVAL_UFLOW_NONNAN instead of
DBL_NARROW_EVAL, reloading the PIC register as needed.
* sysdeps/i386/fpu/e_powf.S: Use DEFINE_FLT_MIN.
(__ieee754_powf): Use FLT_NARROW_EVAL_UFLOW_NONNAN instead of
FLT_NARROW_EVAL. Use separate return path for case when first
argument is NaN.
* sysdeps/i386/fpu/e_powl.S: Include <i386-math-asm.h>. Use
DEFINE_LDBL_MIN.
(__ieee754_powl): Use LDBL_CHECK_FORCE_UFLOW_NONNAN, reloading the
PIC register.
* sysdeps/ieee754/dbl-64/e_pow.c (__ieee754_pow): Use
math_check_force_underflow_nonneg.
* sysdeps/ieee754/flt-32/e_powf.c (__ieee754_powf): Force
underflow for subnormal result.
* sysdeps/ieee754/ldbl-128/e_powl.c (__ieee754_powl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_powl.c (__ieee754_powl): Use
math_check_force_underflow_nonneg.
* sysdeps/x86/fpu/powl_helper.c (__powl_helper): Use
math_check_force_underflow.
* sysdeps/x86_64/fpu/x86_64-math-asm.h
(LDBL_CHECK_FORCE_UFLOW_NONNAN): New macro.
* sysdeps/x86_64/fpu/e_powl.S: Include <x86_64-math-asm.h>. Use
DEFINE_LDBL_MIN.
(__ieee754_powl): Use LDBL_CHECK_FORCE_UFLOW_NONNAN.
* math/auto-libm-test-in: Add more tests of pow.
* math/auto-libm-test-out: Regenerated.
Similar to various other bugs in this area, hypot functions can fail
to raise the underflow exception when the result is tiny and inexact
but one or more low bits of the intermediate result that is scaled
down (or, in the i386 case, converted from a wider evaluation format)
are zero. This patch forces the exception in a similar way to
previous fixes.
Note that this issue cannot arise for implementations of hypotf using
double (or wider) for intermediate evaluation (if hypotf should
underflow, that means the double square root is being computed of some
number of the form N*2^-298, for 0 < N < 2^46, which is exactly
represented as a double, and whatever the rounding mode such a square
root cannot have a mantissa with all zeroes after the initial 23
bits). Thus no changes are made to hypotf implementations in this
patch, only to hypot and hypotl.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18803]
* sysdeps/i386/fpu/e_hypot.S: Use DEFINE_DBL_MIN.
(MO): New macro.
(__ieee754_hypot) [PIC]: Load PIC register.
(__ieee754_hypot): Use DBL_NARROW_EVAL_UFLOW_NONNEG instead of
DBL_NARROW_EVAL.
* sysdeps/ieee754/dbl-64/e_hypot.c (__ieee754_hypot): Use
math_check_force_underflow_nonneg in case where result might be
tiny.
* sysdeps/ieee754/ldbl-128/e_hypotl.c (__ieee754_hypotl):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_hypotl.c (__ieee754_hypotl):
Likewise.
* sysdeps/ieee754/ldbl-96/e_hypotl.c (__ieee754_hypotl): Likewise.
* sysdeps/powerpc/fpu/e_hypot.c (__ieee754_hypot): Likewise.
* math/auto-libm-test-in: Add more tests of hypot.
* math/auto-libm-test-out: Regenerated.
sysdeps/ieee754/flt-32/e_exp2f.c declares two variable as "static
const volatile float". Maybe this use of "volatile" was originally
intended to inhibit optimization of underflowing / overflowing
operations such as TWOM100 * TWOM100; in any case, it's not currently
needed, as given -frounding-math constant folding of such expressions
is properly disabled when it would be unsafe. This patch removes the
unnecessary use of "volatile".
Tested for x86_64.
* sysdeps/ieee754/flt-32/e_exp2f.c (TWOM100): Remove volatile.
(TWO127): Likewise.
Where glibc code needs to avoid excess range and precision in
floating-point arithmetic, code variously uses either asms or volatile
to force the results of that arithmetic to memory; mostly this is
conditional on FLT_EVAL_METHOD, but in the case of lrint / llrint
functions some use of volatile is unconditional (and is present
unnecessarily in versions for long double). This patch make such code
use the recently-added math_narrow_eval macro consistently, removing
the unnecessary uses of volatile in long double lrint / llrint
implementations completely.
Tested for x86_64, x86, mips64 and powerpc.
* math/s_nexttowardf.c (__nexttowardf): Use math_narrow_eval.
* stdlib/strtod_l.c: Include <math_private.h>.
(overflow_value): Use math_narrow_eval.
(underflow_value): Likewise.
* sysdeps/i386/fpu/s_nexttoward.c (__nexttoward): Likewise.
* sysdeps/i386/fpu/s_nexttowardf.c (__nexttowardf): Likewise.
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Likewise.
(__ieee754_gamma_r): Likewise.
* sysdeps/ieee754/dbl-64/gamma_productf.c (__gamma_productf):
Likewise.
* sysdeps/ieee754/dbl-64/k_rem_pio2.c (__kernel_rem_pio2):
Likewise.
* sysdeps/ieee754/dbl-64/lgamma_neg.c (__lgamma_neg): Likewise.
* sysdeps/ieee754/dbl-64/s_erf.c (__erfc): Likewise.
* sysdeps/ieee754/dbl-64/s_llrint.c (__llrint): Likewise.
* sysdeps/ieee754/dbl-64/s_lrint.c (__lrint): Likewise.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Likewise.
(__ieee754_gammaf_r): Likewise.
* sysdeps/ieee754/flt-32/k_rem_pio2f.c (__kernel_rem_pio2f):
Likewise.
* sysdeps/ieee754/flt-32/lgamma_negf.c (__lgamma_negf): Likewise.
* sysdeps/ieee754/flt-32/s_erff.c (__erfcf): Likewise.
* sysdeps/ieee754/flt-32/s_llrintf.c (__llrintf): Likewise.
* sysdeps/ieee754/flt-32/s_lrintf.c (__lrintf): Likewise.
* sysdeps/ieee754/ldbl-128/s_llrintl.c (__llrintl): Do not use
volatile.
* sysdeps/ieee754/ldbl-128/s_lrintl.c (__lrintl): Likewise.
* sysdeps/ieee754/ldbl-128/s_nexttoward.c (__nexttoward): Use
math_narrow_eval.
* sysdeps/ieee754/ldbl-128ibm/s_nexttoward.c (__nexttoward):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-96/gamma_product.c (__gamma_product):
Likewise.
* sysdeps/ieee754/ldbl-96/s_llrintl.c (__llrintl): Do not use
volatile.
* sysdeps/ieee754/ldbl-96/s_lrintl.c (__lrintl): Likewise.
* sysdeps/ieee754/ldbl-96/s_nexttoward.c (__nexttoward): Use
math_narrow_eval.
* sysdeps/ieee754/ldbl-96/s_nexttowardf.c (__nexttowardf):
Likewise.
* sysdeps/ieee754/ldbl-opt/s_nexttowardfd.c (__nldbl_nexttowardf):
Likewise.
Various i386 libm functions return values with excess range and
precision; Wilco Dijkstra's patches to make isfinite etc. expand
inline cause this pre-existing issue to result in test failures (when
e.g. a result that overflows float but not long double gets counted as
overflowing for some purposes but not others).
This patch addresses those cases arising from functions defined in C,
adding a math_narrow_eval macro that forces values to memory to
eliminate excess precision if FLT_EVAL_METHOD indicates this is
needed, and is a no-op otherwise. I'll convert existing uses of
volatile and asm for this purpose to use the new macro later, once
i386 has clean test results again (which requires fixes for .S files
as well).
Tested for x86_64 and x86. Committed.
[BZ #18980]
* sysdeps/generic/math_private.h: Include <float.h>.
(math_narrow_eval): New macro.
[FLT_EVAL_METHOD != 0] (excess_precision): Likewise.
* sysdeps/ieee754/dbl-64/e_cosh.c (__ieee754_cosh): Use
math_narrow_eval on overflowing return value.
* sysdeps/ieee754/dbl-64/e_lgamma_r.c (__ieee754_lgamma_r):
Likewise.
* sysdeps/ieee754/dbl-64/e_sinh.c (__ieee754_sinh): Likewise.
* sysdeps/ieee754/flt-32/e_coshf.c (__ieee754_coshf): Likewise.
* sysdeps/ieee754/flt-32/e_lgammaf_r.c (__ieee754_lgammaf_r):
Likewise.
* sysdeps/ieee754/flt-32/e_sinhf.c (__ieee754_sinhf): Likewise.
Bug 15384 notes that in __finite, two different constants are used
that could be the same constant (the result only depends on the
exponent of the floating-point representation), and that using the
same constant is better for architectures where constants need loading
from a constant pool. This patch implements that change.
Tested for x86_64, mips64 and powerpc.
[BZ #15384]
* sysdeps/ieee754/dbl-64/s_finite.c (FINITE): Use same constant as
bit-mask as in subtraction.
* sysdeps/ieee754/dbl-64/wordsize-64/s_finite.c (__finite):
Likewise.
* sysdeps/ieee754/flt-32/s_finitef.c (FINITEF): Likewise.
* sysdeps/ieee754/ldbl-128/s_finitel.c (__finitel): Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_finitel.c (__finitel): Likewise.
Similar to various other bugs in this area, tgamma functions can fail
to raise the underflow exception when the result is tiny and inexact
but one or more low bits of the intermediate result that is scaled
down are zero. This patch forces the exception in a similar way to
previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18951]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (__ieee754_gamma_r): Force
underflow exception for small results.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (__ieee754_gammaf_r):
Likewise.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (__ieee754_gammal_r):
Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (__ieee754_gammal_r):
Likewise.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (__ieee754_gammal_r):
Likewise.
* math/auto-libm-test-in: Add more tests of tgamma.
* math/auto-libm-test-out: Regenerated.
As noted in bug 6803, scalbn fails to set errno on overflow and
underflow. This patch fixes this by making scalbn an alias of ldexp,
which has exactly the same semantics (for floating-point types with
radix 2) and already has wrappers that deal with setting errno,
instead of an alias of the internal __scalbn (which ldexp calls).
Notes:
* Where compat symbols were defined for scalbn functions, I didn't
change what they point to (to keep the patch minimal), so such
compat symbols continue to go directly to the non-errno-setting
functions.
* Mike, I didn't do anything with the IA64 versions of these
functions, where I think both the ldexp and scalbn functions already
deal with setting errno. As a cleanup (not needed to fix this bug)
however you might want to make those functions into aliases for
IA64; there is no need for them to be separate function
implementations at all.
* This concludes the fix for bug 6803 since the scalb and scalbln
cases of that bug were fixed some time ago.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #6803]
* math/s_ldexp.c (scalbn): Define as weak alias of __ldexp.
[NO_LONG_DOUBLE] (scalbnl): Define as weak alias of __ldexp.
* math/s_ldexpf.c (scalbnf): Define as weak alias of __ldexpf.
* math/s_ldexpl.c (scalbnl): Define as weak alias of __ldexpl.
* sysdeps/i386/fpu/s_scalbn.S (scalbn): Remove alias.
* sysdeps/i386/fpu/s_scalbnf.S (scalbnf): Likewise.
* sysdeps/i386/fpu/s_scalbnl.S (scalbnl): Likewise.
* sysdeps/ieee754/dbl-64/s_scalbn.c (scalbn): Likewise.
[NO_LONG_DOUBLE] (scalbnl): Likewise.
* sysdeps/ieee754/dbl-64/wordsize-64/s_scalbn.c (scalbn):
Likewise.
[NO_LONG_DOUBLE] (scalbnl): Likewise.
* sysdeps/ieee754/flt-32/s_scalbnf.c (scalbnf): Likewise.
* sysdeps/ieee754/ldbl-128/s_scalbnl.c (scalbnl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/s_scalbnl.c (scalbnl): Remove
long_double_symbol calls.
* sysdeps/ieee754/ldbl-64-128/s_scalbnl.c (scalbnl): Likewise.
* sysdeps/ieee754/ldbl-opt/s_ldexpl.c (__ldexpl_2): Define as
strong alias of __ldexpl.
(scalbnl): Define using long_double_symbol.
* sysdeps/m68k/m680x0/fpu/s_scalbn.c (__CONCATX(scalbn,suffix)):
Remove alias.
* sysdeps/sparc/sparc64/soft-fp/s_scalbnl.c (scalbnl): Likewise.
* sysdeps/x86_64/fpu/s_scalbnl.S (scalbnl): Likewise.
* math/libm-test.inc (scalbn_test_data): Add errno expectations.
(scalbln_test_data): Add more errno expectations.
The ldbl-128 and ldbl-128ibm expm1l implementations have code to
handle +Inf and finite arguments above an overflow threshold. Since
they now use __expl for large positive arguments to fix other
problems, this code is unreachable; this patch removes it.
Tested for mips64 and powerpc.
[BZ #16415]
* sysdeps/ieee754/ldbl-128/s_expm1l.c (maxlog): Remove variable.
(__expm1l): Remove code to handle positive infinity and overflow.
* sysdeps/ieee754/ldbl-128ibm/s_expm1l.c (maxlog): Remove
variable.
(__expm1l): Remove code to handle positive infinity and overflow.
The ldbl-128ibm implementation of nearbyintl wrongly uses signaling
comparisons such as "if (fabs (u.d[0].d) < TWO52)" on arguments that
might be NaNs, when "invalid" exceptions should not be raised. (For
hard float, this issue may be hidden by
<https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58684>, powerpc GCC
wrongly only using unordered comparison instructions.) This patch
fixes this by just returning the argument if it is not finite (because
of the arbitrary value of the low part of a NaN in IBM long double,
there are quite a lot of comparisons that could end up involving a NaN
when the argument to nearbyintl is a NaN, so excluding NaN arguments
at the start is the simplest and safest fix).
Tested for powerpc-nofpu, where it removes failures for spurious
"invalid" exceptions from nearbyintl.
[BZ #18857]
* sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c (__nearbyintl): Just
return non-finite argument without doing ordered comparisons on
it.
Bug 15918 points out that the handling of infinities in hypotf can be
simplified: it's enough to return the absolute value of the infinite
argument without first comparing it to the other argument and possibly
returning that other argument's absolute value. This patch makes that
cleanup (which should not change how hypotf behaves on any input).
Tested for x86_64.
[BZ #15918]
* sysdeps/ieee754/flt-32/e_hypotf.c (__ieee754_hypotf): Simplify
handling of cases where one argument is an infinity.
Various exp2 implementations in glibc can miss underflow exceptions
when the scaling down part of the calculation is exact (or, in the x86
case, when the conversion from extended precision to the target
precision is exact). This patch forces the exception in a similar way
to previous fixes.
The x86 exp2f changes may in fact not be needed for this purpose -
it's likely to be the case that no argument of type float has an exp2
result so close to an exact subnormal float value that it equals that
value when rounded to 64 bits (even taking account of variation
between different x86 implementations). However, they are included
for consistency with the changes to exp2 and so as to fix the exp2f
part of bug 18875 by ensuring that excess range and precision is
removed from underflowing return values.
Tested for x86_64, x86 and mips64.
[BZ #16521]
[BZ #18875]
* math/e_exp2l.c (__ieee754_exp2l): Force underflow exception for
small results.
* sysdeps/i386/fpu/e_exp2.S (dbl_min): New object.
(MO): New macro.
(__ieee754_exp2): For small results, force underflow exception and
remove excess range and precision from return value.
* sysdeps/i386/fpu/e_exp2f.S (flt_min): New object.
(MO): New macro.
(__ieee754_exp2f): For small results, force underflow exception
and remove excess range and precision from return value.
* sysdeps/i386/fpu/e_exp2l.S (ldbl_min): New object.
(MO): New macro.
(__ieee754_exp2l): Force underflow exception for small results.
* sysdeps/ieee754/dbl-64/e_exp2.c (__ieee754_exp2): Likewise.
* sysdeps/ieee754/flt-32/e_exp2f.c (__ieee754_exp2f): Likewise.
* sysdeps/x86_64/fpu/e_exp2l.S (ldbl_min): New object.
(MO): New macro.
(__ieee754_exp2l): Force underflow exception for small results.
* math/auto-libm-test-in: Add more tests or exp2.
* math/auto-libm-test-out: Regenerated.
The ldbl-128 / ldbl-128ibm implementation of lgammal converts (the
floor of minus) non-integer negative arguments to int to determine the
value of signgam. When those values are outside the range of int,
this produces spurious "invalid" exceptions and incorrect values of
signgam. This patch fixes this by instead determining signgam through
comparing half the integer in question to floor of half the integer.
Tested for mips64, x86_64 and x86.
[BZ #18952]
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c (__ieee754_lgammal_r): Do
not convert non-integer negative arguments to int to determine the
value of signgam.
* math/auto-libm-test-in: Add more tests of lgamma.
* math/auto-libm-test-out: Regenerated.
The existing implementations of lgamma functions (except for the ia64
versions) use the reflection formula for negative arguments. This
suffers large inaccuracy from cancellation near zeros of lgamma (near
where the gamma function is +/- 1).
This patch fixes this inaccuracy. For arguments above -2, there are
no zeros and no large cancellation, while for sufficiently large
negative arguments the zeros are so close to integers that even for
integers +/- 1ulp the log(gamma(1-x)) term dominates and cancellation
is not significant. Thus, it is only necessary to take special care
about cancellation for arguments around a limited number of zeros.
Accordingly, this patch uses precomputed tables of relevant zeros,
expressed as the sum of two floating-point values. The log of the
ratio of two sines can be computed accurately using log1p in cases
where log would lose accuracy. The log of the ratio of two gamma(1-x)
values can be computed using Stirling's approximation (the difference
between two values of that approximation to lgamma being computable
without computing the two values and then subtracting), with
appropriate adjustments (which don't reduce accuracy too much) in
cases where 1-x is too small to use Stirling's approximation directly.
In the interval from -3 to -2, using the ratios of sines and of
gamma(1-x) can still produce too much cancellation between those two
parts of the computation (and that interval is also the worst interval
for computing the ratio between gamma(1-x) values, which computation
becomes more accurate, while being less critical for the final result,
for larger 1-x). Because this can result in errors slightly above
those accepted in glibc, this interval is instead dealt with by
polynomial approximations. Separate polynomial approximations to
(|gamma(x)|-1)(x-n)/(x-x0) are used for each interval of length 1/8
from -3 to -2, where n (-3 or -2) is the nearest integer to the
1/8-interval and x0 is the zero of lgamma in the relevant half-integer
interval (-3 to -2.5 or -2.5 to -2).
Together, the two approaches are intended to give sufficient accuracy
for all negative arguments in the problem range. Outside that range,
the previous implementation continues to be used.
Tested for x86_64, x86, mips64 and powerpc. The mips64 and powerpc
testing shows up pre-existing problems for ldbl-128 and ldbl-128ibm
with large negative arguments giving spurious "invalid" exceptions
(exposed by newly added tests for cases this patch doesn't affect the
logic for); I'll address those problems separately.
[BZ #2542]
[BZ #2543]
[BZ #2558]
* sysdeps/ieee754/dbl-64/e_lgamma_r.c (__ieee754_lgamma_r): Call
__lgamma_neg for arguments from -28.0 to -2.0.
* sysdeps/ieee754/flt-32/e_lgammaf_r.c (__ieee754_lgammaf_r): Call
__lgamma_negf for arguments from -15.0 to -2.0.
* sysdeps/ieee754/ldbl-128/e_lgammal_r.c (__ieee754_lgammal_r):
Call __lgamma_negl for arguments from -48.0 or -50.0 to -2.0.
* sysdeps/ieee754/ldbl-96/e_lgammal_r.c (__ieee754_lgammal_r):
Call __lgamma_negl for arguments from -33.0 to -2.0.
* sysdeps/ieee754/dbl-64/lgamma_neg.c: New file.
* sysdeps/ieee754/dbl-64/lgamma_product.c: Likewise.
* sysdeps/ieee754/flt-32/lgamma_negf.c: Likewise.
* sysdeps/ieee754/flt-32/lgamma_productf.c: Likewise.
* sysdeps/ieee754/ldbl-128/lgamma_negl.c: Likewise.
* sysdeps/ieee754/ldbl-128/lgamma_productl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/lgamma_negl.c: Likewise.
* sysdeps/ieee754/ldbl-128ibm/lgamma_productl.c: Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_negl.c: Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_product.c: Likewise.
* sysdeps/ieee754/ldbl-96/lgamma_productl.c: Likewise.
* sysdeps/generic/math_private.h (__lgamma_negf): New prototype.
(__lgamma_neg): Likewise.
(__lgamma_negl): Likewise.
(__lgamma_product): Likewise.
(__lgamma_productl): Likewise.
* math/Makefile (libm-calls): Add lgamma_neg and lgamma_product.
* math/auto-libm-test-in: Add more tests of lgamma.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
As with other spots in the code, GCC 4.8 unnecessarily complains about
an uninitialized variable in tanl calcs, so this patch disables. With
it, the library and sees the usual set of test passes.
* sysdeps/ieee754/ldbl-96/k_tanl.c: Include <libc-internal.h>.
(__kernel_tanl): Ignore uninitialized warnings around use of SIGN.
Removing the use of -Wno-uninitialized for math/ shows errors for
ldbl-128ibm:
../sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c: In function '__nearbyintl':
../sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c:119:34: error: 'low' may be used uninitialized in this function [-Werror=maybe-uninitialized]
u.d[1].d = high - u.d[0].d + low;
^
../sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c:119:23: error: 'high' may be used uninitialized in this function [-Werror=maybe-uninitialized]
u.d[1].d = high - u.d[0].d + low;
^
These errors are correct: if the high part of the argument is a NaN,
and the low part is nonzero but has absolute value less than 2^52,
those variables can be used uninitialized. This patch rearranges the
code so that the variables are always initialized with the natural
values, and then possibly modified later, to avoid this uninitialized
use. (Note that there are still other issues with this code and NaNs
that are not fixed by this patch.) No bug filed in Bugzilla or
testcase added for the uninitialized use since it wasn't user-visible
with the compiler I tried (that is, I still got a NaN result).
Tested for powerpc.
* sysdeps/ieee754/ldbl-128ibm/s_nearbyintl.c: Always initialize
variables for high and low parts before possibly modifying them.
Various fma implementations have logic that, when computing fma (x, y,
z) where z is large (so care needs taking to avoid internal overflow)
but x * y is small, scale x * y up instead of down to avoid internal
underflows resulting from scaling down. (In these cases, x * y is
small enough that only its sign actually matters rather than the exact
value.)
The threshold for scaling up instead of down was correct for "if the
unscaled values were multiplied, the low part of the multiplication
could underflow", and the scaling was sufficient to ensure that the
low part of the multiplication did not underflow (given that cases of
very small x * y - less than half the least subnormal - were
previously dealt with). However, the choice in the functions wasn't
between scaling up or no scaling, but between scaling up and scaling
down (scaling down actually being needed when x * y isn't so small
compared to z and so the exact value does matter). Thus a larger
threshold is needed to ensure that scaling down doesn't produce values
the multiplication of whose low parts underflows. This patch
increases the thresholds accordingly.
Tested for x86_64, x86 and mips64 (with the MIPS version of s_fmal.c
removed so that the ldbl-128 version gets tested instead of the
soft-fp one).
[BZ #18824]
* sysdeps/ieee754/dbl-64/s_fma.c (__fma): Increase threshold for
scaling x * y up instead of down.
* sysdeps/ieee754/ldbl-128/s_fmal.c (__fmal): Likewise.
* sysdeps/ieee754/ldbl-96/s_fmal.c (__fmal): Likewise.
* math/auto-libm-test-in: Add more tests of fma.
* math/auto-libm-test-out: Regenerated.
Similar to various other bugs in this area, some tanh implementations
do not raise the underflow exception for subnormal arguments, when the
result is tiny and inexact. This patch forces the exception in a
similar way to previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16520]
* sysdeps/ieee754/dbl-64/s_tanh.c: Include <float.h>.
(__tanh): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/flt-32/s_tanhf.c: Include <float.h>.
(__tanhf): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/ldbl-128/s_tanhl.c: Include <float.h>.
(__tanhl): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/ldbl-128ibm/s_tanhl.c: Include <float.h>.
(__tanhl): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/ldbl-96/s_tanhl.c: Include <float.h>.
(__tanhl): Force underflow exception for arguments with small
absolute value.
* math/auto-libm-test-in: Add more tests of tanh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
ldbl-128ibm tanhl uses a too-small threshold to decide when to return
+/-1, resulting in large errors. This patch changes it to a more
appropriate threshold (the requirement is for 2*exp(-2|x|) to be small
in terms of ulps of 1).
Tested for x86_64, x86 and powerpc.
[BZ #18790]
* sysdeps/ieee754/ldbl-128ibm/s_tanhl.c (__tanhl): Increase
threshold for returning +/- 1.
* math/auto-libm-test-in: Add more tests of tanh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
ldbl-128ibm sinhl uses a too-big threshold to decide when to return
the argument, resulting in large errors. This patch fixes it to use a
more appropriate threshold.
Tested for x86_64, x86 and powerpc.
[BZ #18789]
* sysdeps/ieee754/ldbl-128ibm/e_sinhl.c (__ieee754_sinhl): Use
smaller threshold for returning the argument.
* math/auto-libm-test-in: Add more tests of sinh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
Similar to various other bugs in this area, some tan implementations
do not raise the underflow exception for subnormal arguments, when the
result is tiny and inexact. This patch forces the exception in a
similar way to previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16517]
* sysdeps/ieee754/dbl-64/s_tan.c: Include <float.h>.
(tan): Force underflow exception for arguments with small absolute
value.
* sysdeps/ieee754/flt-32/k_tanf.c: Include <float.h>.
(__kernel_tanf): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128/k_tanl.c: Include <float.h>.
(__kernel_tanl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128ibm/k_tanl.c: Include <float.h>.
(__kernel_tanl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-96/k_tanl.c: Include <float.h>.
(__kernel_tanl): Force underflow exception for arguments with
small absolute value.
* math/auto-libm-test-in: Add more tests of tan.
* math/auto-libm-test-out: Regenerated.
Similar to various other bugs in this area, some sinh implementations
do not raise the underflow exception for subnormal arguments, when the
result is tiny and inexact. This patch forces the exception in a
similar way to previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16519]
* sysdeps/ieee754/dbl-64/e_sinh.c: Include <float.h>.
(__ieee754_sinh): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/flt-32/e_sinhf.c: Include <float.h>.
(__ieee754_sinhf): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128/e_sinhl.c: Include <float.h>.
(__ieee754_sinhl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128ibm/e_sinhl.c: Include <float.h>.
(__ieee754_sinhl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-96/e_sinhl.c: Include <float.h>.
(__ieee754_sinhl): Force underflow exception for arguments with
small absolute value.
* math/auto-libm-test-in: Add more tests of sinh.
* math/auto-libm-test-out: Regenerated.
* sysdeps/i386/fpu/libm-test-ulps: Update.
The flt-32 implementation of powf wrongly uses x-1 instead of |x|-1
when computing log (x) for the case where |x| is close to 1 and y is
large. This patch fixes the logic accordingly. Relevant tests
existed for x close to 1, and corresponding tests are added for x
close to -1, as well as for some new variant cases.
Tested for x86_64 and x86.
[BZ #18647]
* sysdeps/ieee754/flt-32/e_powf.c (__ieee754_powf): For large y
and |x| close to 1, use absolute value of x when computing log.
* math/auto-libm-test-in: Add more tests of pow.
* math/auto-libm-test-out: Regenerated.
In the ldbl-128 implementation of expm1l, when expm1l's result should
underflow to 0 (argument minus the least subnormal, in some rounding
modes), it can be a zero of the wrong sign. This patch fixes this in
the same way previously used for the x86 / x86_64 versions.
Tested for mips64.
[BZ #18619]
* sysdeps/ieee754/ldbl-128/s_expm1l.c (__expm1l): Force underflow
and return argument in case of subnormal argument.
In non-default rounding modes, tgamma can be slightly less accurate
than permitted by glibc's accuracy goals.
Part of the problem is error accumulation, addressed in this patch by
setting round-to-nearest for internal computations. However, there
was also a bug in the code dealing with computing pow (x + n, x + n)
where x + n is not exactly representable, providing another source of
error even in round-to-nearest mode; it was necessary to address both
bugs to get errors for all testcases within glibc's accuracy goals.
Given this second fix, accuracy in round-to-nearest mode is also
improved (hence regeneration of ulps for tgamma should be from scratch
- truncate libm-test-ulps or at least remove existing tgamma entries -
so that the expected ulps can be reduced).
Some additional complications also arose. Certain tgamma tests should
strictly, according to IEEE semantics, overflow or not depending on
the rounding mode; this is beyond the scope of glibc's accuracy goals
for any function without exactly-determined results, but
gen-auto-libm-tests doesn't handle being lax there as it does for
underflow. (libm-test.inc also doesn't handle being lax about whether
the result in cases very close to the overflow threshold is infinity
or a finite value close to overflow, but that doesn't cause problems
in this case though I've seen it cause problems with random test
generation for some functions.) Thus, spurious-overflow markings,
with a comment, are added to auto-libm-test-in (no bug in Bugzilla
because the issue is with the testsuite, not a user-visible bug in
glibc). And on x86, after the patch I saw ERANGE issues as previously
reported by Carlos (see my commentary in
<https://sourceware.org/ml/libc-alpha/2015-01/msg00485.html>), which
needed addressing by ensuring excess range and precision were
eliminated at various points if FLT_EVAL_METHOD != 0.
I also noticed and fixed a cosmetic issue where 1.0f was used in long
double functions and should have been 1.0L.
This completes the move of all functions to testing in all rounding
modes with ALL_RM_TEST, so gen-libm-have-vector-test.sh is updated to
remove the workaround for some functions not using ALL_RM_TEST.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #18613]
* sysdeps/ieee754/dbl-64/e_gamma_r.c (gamma_positive): Take log of
X_ADJ not X when adjusting exponent.
(__ieee754_gamma_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/flt-32/e_gammaf_r.c (gammaf_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammaf_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed.
* sysdeps/ieee754/ldbl-128/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-128ibm/e_gammal_r.c (gammal_positive): Take
log of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* sysdeps/ieee754/ldbl-96/e_gammal_r.c (gammal_positive): Take log
of X_ADJ not X when adjusting exponent.
(__ieee754_gammal_r): Do intermediate computations in
round-to-nearest then adjust overflowing and underflowing results
as needed. Use 1.0L not 1.0f as numerator of division.
* math/libm-test.inc (tgamma_test_data): Remove one test. Moved
to auto-libm-test-in.
(tgamma_test): Use ALL_RM_TEST.
* math/auto-libm-test-in: Add one test of tgamma. Mark some other
tests of tgamma with spurious-overflow.
* math/auto-libm-test-out: Regenerated.
* math/gen-libm-have-vector-test.sh: Do not check for START.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
The ldbl-128 implementation of j1l produces spurious underflow
exceptions for some small arguments, as a result of squaring the
argument. This patch fixes it just to use a linear approximation for
sufficiently small arguments, and then to force an underflow exception
only in the cases where it is required.
Tested for mips64.
[BZ #18612]
* sysdeps/ieee754/ldbl-128/e_j1l.c (__ieee754_j1l): For small
arguments, just return 0.5 times the argument, with underflow
forced as needed.
* math/auto-libm-test-in: Add more tests of j1.
* math/auto-libm-test-out: Regenerated.
Similar to various other bugs in this area, j1 and jn implementations
can fail to raise the underflow exception when the internal
computation is exact although the actual function is inexact. This
patch forces the exception in a similar way to other such fixes. (The
ldbl-128 / ldbl-128ibm j1l implementation is different and doesn't
need a change for this until spurious underflows in it are fixed.)
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16559]
* sysdeps/ieee754/dbl-64/e_j1.c: Include <float.h>.
(__ieee754_j1): Force underflow exception for small results.
* sysdeps/ieee754/dbl-64/e_jn.c (__ieee754_jn): Likewise.
* sysdeps/ieee754/flt-32/e_j1f.c: Include <float.h>.
(__ieee754_j1f): Force underflow exception for small results.
* sysdeps/ieee754/flt-32/e_jnf.c (__ieee754_jnf): Likewise.
* sysdeps/ieee754/ldbl-128/e_jnl.c (__ieee754_jnl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_jnl.c (__ieee754_jnl): Likewise.
* sysdeps/ieee754/ldbl-96/e_j1l.c: Include <float.h>.
(__ieee754_j1l): Force underflow exception for small results.
* sysdeps/ieee754/ldbl-96/e_jnl.c (__ieee754_jnl): Likewise.
* math/auto-libm-test-in: Add more tests of j1 and jn.
* math/auto-libm-test-out: Regenerated.
Some existing jn tests, if run in non-default rounding modes, produce
errors above those accepted in glibc, which causes problems for moving
tests of jn to use ALL_RM_TEST. This patch makes jn set rounding
to-nearest internally, as was done for yn some time ago, then computes
the appropriate underflowing value for results that underflowed to
zero in to-nearest, and moves the tests to ALL_RM_TEST. It does
nothing about the general inaccuracy of Bessel function
implementations in glibc, though it should make jn more accurate on
average in non-default rounding modes through reduced error
accumulation. The recomputation of results that underflowed to zero
should as a side-effect fix some cases of bug 16559, where jn just
used an exact zero, but that is *not* the goal of this patch and other
cases of that bug remain unfixed.
(Most of the changes in the patch are reindentation to add new scopes
for SET_RESTORE_ROUND*.)
Tested for x86_64, x86, powerpc and mips64.
[BZ #16559]
[BZ #18602]
* sysdeps/ieee754/dbl-64/e_jn.c (__ieee754_jn): Set
round-to-nearest internally then recompute results that
underflowed to zero in the original rounding mode.
* sysdeps/ieee754/flt-32/e_jnf.c (__ieee754_jnf): Likewise.
* sysdeps/ieee754/ldbl-128/e_jnl.c (__ieee754_jnl): Likewise.
* sysdeps/ieee754/ldbl-128ibm/e_jnl.c (__ieee754_jnl): Likewise.
* sysdeps/ieee754/ldbl-96/e_jnl.c (__ieee754_jnl): Likewise
* math/libm-test.inc (jn_test): Use ALL_RM_TEST.
* sysdeps/i386/fpu/libm-test-ulps: Update.
* sysdeps/x86_64/fpu/libm-test-ulps: Likewise.
Similar to various other bugs in this area, the ldbl-128 expl
implementation does not raise the underflow exception for all
subnormal results, if the scaling down is exact although the actual
result is inexact. This patch fixes this by forcing the exception in
this case (the tests that failed before and pass after the test are
already in the testsuite).
Tested for mips64.
[BZ #18586]
* sysdeps/ieee754/ldbl-128/e_expl.c (__ieee754_expl): Force
underflow exception for small results.
Similar to various other bugs in this area, some sin and sincos
implementations do not raise the underflow exception for subnormal
arguments, when the result is tiny and inexact. This patch forces the
exception in a similar way to previous fixes.
Tested for x86_64, x86, mips64 and powerpc.
[BZ #16526]
[BZ #16538]
* sysdeps/ieee754/dbl-64/s_sin.c: Include <float.h>.
(__sin): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/flt-32/k_sinf.c: Include <float.h>.
(__kernel_sinf): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128/k_sincosl.c: Include <float.h>.
(__kernel_sincosl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128/k_sinl.c: Include <float.h>.
(__kernel_sinl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128ibm/k_sincosl.c: Include <float.h>.
(__kernel_sincosl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-128ibm/k_sinl.c: Include <float.h>.
(__kernel_sinl): Force underflow exception for arguments with
small absolute value.
* sysdeps/ieee754/ldbl-96/k_sinl.c: Include <float.h>.
(__kernel_sinl): Force underflow exception for arguments with
small absolute value.
* sysdeps/powerpc/fpu/k_sinf.c: Include <float.h>.
(__kernel_sinf): Force underflow exception for arguments with
small absolute value.
* math/auto-libm-test-in: Add more tests of sin and sincos.
* math/auto-libm-test-out: Regenerated.
__kernel_standard_l converts long double arguments to double for use
in SVID "struct exception". This has special-case handling for when
that conversion would overflow or underflow but the original long
double function wouldn't. However, it turns out that "inexact"
exceptions can be spurious here as well, when the function is exactly
determined and __kernel_standard_l is being called for a domain error.
This patch fixes this by using feholdexcept / fesetenv to avoid
exceptions from the conversion, replacing the previous special-case
logic for overflow and underflow (this covers all functions using
__kernel_standard_l, not just those that actually need a change, since
there doesn't seem to be much point in restricting things just to the
functions that mustn't get "inexact" here).
Tested for x86_64 and x86.
[BZ #18245]
[BZ #18583]
* sysdeps/ieee754/k_standardl.c: Include <fenv.h>.
(__kernel_standard_l): Use feholdexcept and fesetenv around
conversion to double instead of special-casing overflow and
underflow.
* math/libm-test.inc (fmod_test_data): Add more tests.
(remainder_test_data): Likewise.
(sqrt_test_data): Likewise.
The dbl-64 and flt-32 implementations of exp2 functions produce
spurious underflow exceptions. The underlying reason is the same in
both cases: the computation works as (2^a - 1)*2^b + 2^b for suitably
chosen a and b, where a has small magnitude so 2^a - 1 can be computed
with a low-degree polynomial approximation, and (2^a - 1)*2^b can
underflow even when the final result does not. This patch fixes this
by adjusting the threshold for when scaling is used to avoid
intermediate underflow so it works for any possible value of a where
the final result would not underflow.
Tested for x86_64 and x86.
[BZ #18219]
* sysdeps/ieee754/dbl-64/e_exp2.c (__ieee754_exp2): Reduce
threshold on absolute value of exponent for which scaling is used.
* sysdeps/ieee754/flt-32/e_exp2f.c (__ieee754_exp2f): Likewise.
* math/auto-libm-test-in: Add more tests of exp2.
* math/auto-libm-test-out: Regenerated.
Similar to various other bugs in this area, some expm1 implementations
do not raise the underflow exception for subnormal arguments, when the
result is tiny and inexact. This patch forces the exception in a
similar way to previous fixes.
(The issue does not apply to the ldbl-* implementations or to those
for x86 / x86_64 long double. The change to
sysdeps/ieee754/dbl-64/wordsize-64/e_cosh.c is one I missed when
previously fixing bug 16354; the bug in that implementation was
previously latent, but the expm1 fixes stopped it being latent and so
required it to be fixed to avoid spurious underflows from cosh.)
Tested for x86_64 and x86.
[BZ #16353]
* sysdeps/i386/fpu/s_expm1.S (dbl_min): New object.
(__expm1): Force underflow exception for arguments with small
absolute value.
* sysdeps/i386/fpu/s_expm1f.S (flt_min): New object.
(__expm1f): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/dbl-64/s_expm1.c: Include <float.h>.
(__expm1): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/flt-32/s_expm1f.c: Include <float.h>.
(__expm1f): Force underflow exception for arguments with small
absolute value.
* sysdeps/ieee754/dbl-64/wordsize-64/e_cosh.c (__ieee754_cosh):
Check for small arguments before calling __expm1.
* math/auto-libm-test-in: Do not mark underflow exceptions as
possibly missing for bug 16353.
* math/auto-libm-test-out: Regenerated.
