summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* [TODO] Updated URL of TS 18661-4.vlefevre2015-06-281-1/+1
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9601 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/frexp.c] Handle internal overflow.vlefevre2015-06-261-1/+29
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9600 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfrexp.c] Forgot a "exit (1);".vlefevre2015-06-261-0/+1
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9599 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tmul_2exp.c] Overflow test: more output info; improved code.vlefevre2015-06-261-11/+14
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9598 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/mul_2si.c] Fixed overflow case when n = 0.vlefevre2015-06-262-11/+13
| | | | | | | [tests/tmul_2exp.c] In the overflow test, check mpfr_mul_2ui and mpfr_mul_2si too (mpfr_mul_2ui was OK). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9597 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/div_2si.c] Fixed overflow case when n = 0.vlefevre2015-06-262-21/+28
| | | | | | [tests/tmul_2exp.c] In the overflow test, check mpfr_div_2si too. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9596 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/div_2ui.c] Fixed overflow case (can occur only when n = 0).vlefevre2015-06-262-6/+34
| | | | | | [tests/tmul_2exp.c] Completed the overflow test. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9595 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfrexp.c] Added tests (currently fail due to the mpfr_div_2uivlefevre2015-06-261-0/+112
| | | | | | unhandled-overflow bug). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9594 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tmul_2exp.c] Added tests where mpfr_div_2ui overflows, whichvlefevre2015-06-261-0/+35
| | | | | | is currently not handled. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9593 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/frexp.c] Fixed a bug occurring when the current exponent rangevlefevre2015-06-261-1/+2
| | | | | | does not contain 0. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9592 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/check.c] Typo in a comment.vlefevre2015-06-251-1/+1
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9591 280ebfd0-de03-0410-8827-d642c229c3f4
* [tools/announce-text] Update due to recent InriaForge upgrade.vlefevre2015-06-191-2/+3
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9588 280ebfd0-de03-0410-8827-d642c229c3f4
* [doc/mpfr.texi] Added a paragraph about the multiple use of MPFR.vlefevre2015-06-161-0/+11
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9581 280ebfd0-de03-0410-8827-d642c229c3f4
* [TODO] Added: use symbol versioning.vlefevre2015-06-161-0/+2
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9580 280ebfd0-de03-0410-8827-d642c229c3f4
* [doc/faq.xsl] Remove the indent="no" to get the same behavior as beforevlefevre2015-06-121-2/+1
| | | | | | | when using libxslt 1.1.27+. See: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=718535#10 git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9575 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] Improved the new underflow tests (test_underflow3) tovlefevre2015-06-121-39/+42
| | | | | | test FMA(-x,y,-z). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9572 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/fma.c] Completely fixed the scaling in case of underflow.vlefevre2015-06-121-2/+9
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9569 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/fma.c] Partially fixed the scaling in case of underflow,vlefevre2015-06-121-3/+2
| | | | | | | corresponding to the testcase from r9566. Double rounding is not handled yet (failure triggered by r9567). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9568 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] Improved the new underflow tests (test_underflow3) tovlefevre2015-06-121-37/+41
| | | | | | | | | yield some cases with double rounding when scaling: If the only fix in src/fma.c is to change the rounding mode of the last mpfr_div_2ui (computing s / 2^scale) to rnd_mode, a new failure appears for n = 2, k = 4, s = -1, MPFR_RNDN. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9567 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] Corrected the new underflow tests to really generatevlefevre2015-06-121-8/+8
| | | | | | underflows, showing a bug in mpfr_fma. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9566 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/fma.c] Added log messages.vlefevre2015-06-121-0/+11
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9565 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] The code removed in r9560 was actually necessary,vlefevre2015-06-121-0/+5
| | | | | | but do this in a better way. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9564 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] Added other underflow tests.vlefevre2015-06-111-0/+64
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9563 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] Some improvements.vlefevre2015-06-111-10/+8
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9562 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] GNU coding style.vlefevre2015-06-111-68/+68
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9561 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tfma.c] Removed code that isn't necessary.vlefevre2015-06-111-3/+0
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9560 280ebfd0-de03-0410-8827-d642c229c3f4
* [doc/README.dev] "To make a release": mention -DMPFR_COV_CHECK andvlefevre2015-06-111-7/+14
| | | | | | -DMPFR_ALLOCA_MAX=0 tests. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9557 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/trandom.c] Cleaner test, avoiding the explicit use of mpfr_rands.vlefevre2015-06-101-15/+31
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9553 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/trandom.c] Improved latest change, as some other test alreadyvlefevre2015-06-101-20/+20
| | | | | | | assume a fixed PRNG with GMP >= 4.2.0 (now the value of the last random number of the test is checked). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9551 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/trandom.c]vlefevre2015-06-101-0/+20
| | | | | | | | | | | * Get a non-zero fixed-point number whose first 32 bits are 0 with the default GMP PRNG. This corresponds to the case cnt == 0 && k != 0 in src/urandomb.c (fixed in r8762) with the 32-bit ABI. * Output a warning (with a fatal error when the MPFR_COV_CHECK macro is defined) if one doesn't get such a number, probably because the GMP PRNG has changed. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9550 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/sqrt.c] The bug fixed in r9540 actually came from r9014, whichvlefevre2015-06-101-0/+1
| | | | | | | | incorrectly replaced MPFR_EXP by MPFR_SET_EXP. So, restored the previous code (re-adding the assertion check), keeping my comment from r9540 to explain why we don't use MPFR_SET_EXP. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9545 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tgeneric.c] Fixed overflow/underflow tests introduced in r9539:vlefevre2015-06-101-82/+70
| | | | | | | | | the tests were done in all rounding modes, but the code was incorrect if the exponent of the result depended on the rounding mode. So, the tests are now done only in the rounding mode previously selected. This should be sufficient to detect bugs. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9544 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/lngamma.c] For mpfr_lgamma and mpfr_lngamma, the exponent rangevlefevre2015-06-101-2/+4
| | | | | | was not restored in some cases (some large results). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9543 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/{cos.c,sin.c}] Fixed bug related to the exponent range, introducedvlefevre2015-06-102-3/+8
| | | | | | | | | | | | | in r6461 with the use of mpfr_sincos_fast. Details: For mpfr_sin, the exponent range was not restored, and restoring it before calling mpfr_sincos_fast as this was done for mpfr_cos did not work; indeed the source of this function shows that it needs an extended exponent range. So, changed both mpfr_cos and mpfr_sin to call mpfr_sincos_fast in the extended exponent range and restore the exponent range at the end, like in usual code. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9542 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tgeneric.c] Check whether the exponent range has beenvlefevre2015-06-101-4/+13
| | | | | | | modified by the tested function (which should never happen). This check triggers failures in tlgamma, tlngamma and tsin. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9541 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/sqrt.c] Fixed assertion failure in reduced exponent range.vlefevre2015-06-101-1/+2
| | | | | | | Note: This could happen only in very specific exponent ranges, in particular, which do not include the number 1. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9540 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tgeneric.c] Added overflow/underflow tests (→ 4 failed tests).vlefevre2015-06-101-85/+201
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9539 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/zeta_ui.c]vlefevre2015-06-082-20/+57
| | | | | | | | * Support reduced exponent range for the generic case. * Added logging. [tests/tzeta_ui.c] Added tests in reduced exponent range. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9523 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/add_ui.c] Fixed format string in MPFR_LOG_FUNC.vlefevre2015-06-081-1/+1
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9521 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tzeta_ui.c] Code simplification.vlefevre2015-06-071-10/+8
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9520 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tzeta_ui.c] Test zeta(0) in reduced exponent range and allvlefevre2015-06-071-44/+62
| | | | | | rounding modes. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9519 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/zeta_ui.c] Fixed zeta(0) in reduced exponent range.vlefevre2015-06-072-7/+56
| | | | | | [tests/tzeta_ui.c] Added testcase. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9518 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/{ui_div.c,ui_sub.c}] Correction for reduced exponent range.vlefevre2015-06-052-11/+31
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9514 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tui_sub.c] Check overflow.vlefevre2015-06-051-0/+56
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9513 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tui_div.c] Check overflow.vlefevre2015-06-051-0/+46
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9512 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tgeneric.c] Removed unless parentheses.vlefevre2015-06-051-2/+1
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9511 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tgeneric.c] Special cases with unsigned long argument.vlefevre2015-06-051-9/+9
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9510 280ebfd0-de03-0410-8827-d642c229c3f4
* [tests/tpow_all.c] Better flags output in error message.vlefevre2015-06-051-3/+6
| | | | git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9509 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/set_q.c] Added missing MPFR_SAVE_EXPO_FREE in case of preliminaryvlefevre2015-06-041-0/+2
| | | | | | overflow or underflow (extreme cases). git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9508 280ebfd0-de03-0410-8827-d642c229c3f4
* [src/set_q.c] Replaced a "return" by MPFR_RET: this is at least neededvlefevre2015-06-041-1/+1
| | | | | | in the extreme case cd != 0, and safer in general. git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@9507 280ebfd0-de03-0410-8827-d642c229c3f4
View Lorry Controller Status