summaryrefslogtreecommitdiff
path: root/sysdeps/ieee754/dbl-64
Commit message (Collapse)AuthorAgeFilesLines
* Update copyright notices with scripts/update-copyrights.Joseph Myers2013-01-0285-86/+85
|
* Fix values in __mpexp_twomm1Siddhesh Poyarekar2013-01-021-8/+8
|
* Split mantissa calculation loop and add branch predictionSiddhesh Poyarekar2013-01-021-24/+43
|
* Add assert for potential access beyond array bounds in m1npSiddhesh Poyarekar2013-01-021-4/+17
| | | | | | | | | | | The mpexp code has an access into m1np: for (i=n-1; i>0; i--,n--) { if (m1np[i][p]+m2>0) break; } which could break for p >= 18 or i >= 7. Fortunately this code is never called due to the way the exp function is implemented since values having exponent less than -55 return 1.0. Make sure that if it gets called in future, it is trapped.
* Move more constants into static variablesSiddhesh Poyarekar2013-01-024-22/+8
| | | | Code cleanup.
* Add script to update copyright notices and reformat some to facilitate its use.Joseph Myers2013-01-0123-23/+23
|
* Favour normal numbersSiddhesh Poyarekar2012-12-311-4/+5
|
* Demystify the magic number 134217729.0Siddhesh Poyarekar2012-12-293-8/+10
| | | | | The number 134217729.0 gets used in various places in e_pow.c but there is no explanation of what that number is. Add that explanation.
* Remove unnecessary variable mptwoim1Siddhesh Poyarekar2012-12-282-27/+3
| | | | Code cleanup.
* Replace more constants with their valuesSiddhesh Poyarekar2012-12-285-97/+61
| | | | Code cleanup.
* Replace constants with preprocessor definesSiddhesh Poyarekar2012-12-285-120/+28
| | | | libm Code cleanup.
* Remove redundant __mpexp_nnSiddhesh Poyarekar2012-12-272-28/+2
| | | | It's an array that stores integral float values of the offset.
* Move mpone out to a global constSiddhesh Poyarekar2012-12-277-68/+11
| | | | Code cleanup.
* Fix spurious underflows from pow with results close to 1 (bug 14811).Joseph Myers2012-11-071-0/+4
|
* Fix fma underflows with small x * y (bug 14793).Joseph Myers2012-11-061-18/+27
|
* Fix fma overflow results outside round-to-nearest mode (bug 14797).Joseph Myers2012-11-041-4/+5
|
* Make fma use of Dekker and Knuth algorithms use round-to-nearest (bug 14796).Joseph Myers2012-11-031-2/+16
|
* Fix fma (a, b, c) for small a * b (bugs 14784, 14785).Joseph Myers2012-11-011-4/+32
|
* Fix fma underflow exceptions in after-rounding edge cases.Joseph Myers2012-10-311-0/+12
|
* Fix fma missing underflows and bad results for some subnormal results (bugs ↵Joseph Myers2012-10-301-14/+8
| | | | 14152, 14783).
* Fix sign of inexact zero return from fma (bug 14645).Joseph Myers2012-10-011-0/+5
|
* Fix sign of exact zero return from fma (bug 14638).Joseph Myers2012-09-292-0/+12
|
* Fix inaccuracy of clog, clog10 near |z| = 1 (bug 13629).Joseph Myers2012-09-252-0/+143
|
* Set up errno properly for yn.Marek Polacek2012-07-251-0/+4
|
* Fix wordsize-64 cosh regression (bug 14273).Joseph Myers2012-06-211-0/+1
|
* Fix pow (0.0, -Inf) (bug 14241).Joseph Myers2012-06-151-2/+2
|
* Fix fmod for subnormals (bug 14048).Joseph Myers2012-06-011-2/+2
|
* Ensure additions are not scheduled after fetestexcept in fmaf and fmal.Joseph Myers2012-06-011-1/+3
|
* Add a barrier in the double fma implementation.Aurelien Jarno2012-06-011-1/+3
|
* Optimize handling of denormals in logb/logbf/logblAndreas Schwab2012-05-262-6/+8
|
* Log2 and log10 for wordsize-64.Adhemerval Zanella2012-05-154-82/+305
| | | | This patch also fixes indentation on default dbl-64 code.
* Use __builtin_clzll on int64 in __logbH.J. Lu2012-05-141-1/+1
|
* Fix for logb/logbf/logbl (bugs 13954/13955/13956)Adhemerval Zanella2012-05-102-20/+30
| | | | | | | | POSIX 2008 states that if the input for 'logb[f|l]' is a subnormal number it should be treated as if it were normalized. This means the implementation should calculate the log2 of the mantissa and add it to the subnormal exponent (-126 for float and -1022 for double and IBM long double). This patch takes care of that.
* Fix exp10 inaccuracy and exceptions (bugs 13884, 13914).Joseph Myers2012-05-061-0/+48
|
* Fix missing exceptions from exp (bugs 13787, 13922, 14036).Joseph Myers2012-05-051-16/+5
|
* Remove useless __ilogb*_finite aliasesAndreas Schwab2012-04-181-1/+0
|
* Fix ilogb exception and errno (bug 6794)Adhemerval Zanella2012-04-171-6/+2
| | | | | | | | | | | | | | | | | | | | [BZ #6794] Following Joseph comments about bug 6794, here is a proposed fix. It turned out to be a large fix mainly because I had to move some file along to follow libm files/names conventions. Basically I have added wrappers (w_ilogb.c, w_ilogbf.c, w_ilogbl.c) that now calls the symbol '__ieee754_ilogb'. The wrappers checks for '__ieee754_ilogb' output and set the errno and raise exceptions as expected. The '__ieee754_ilogb' is implemented in sysdeps. I have moved the 's_ilogb[f|l]' files to e_ilogb[f|l] and renamed the '__ilogb[f|l]' to '__ieee754_ilogb[f|l]'. I also found out a bug in i386 and x86-64 assembly coded ilogb implementation where it raises a FE_DIVBYZERO when argument is '0.0'. I corrected this issue as well. Finally I added the errno and FE_INVALID tests for 0.0, NaN and +-InF argument. Tested on i386, x86-64, ppc32 and ppc64.
* Fix missing overflow exceptions from pow (bug 13873).Joseph Myers2012-04-091-2/+3
|
* Use __kernel_standard instead of __kernel_standard_f in exp wrapper (bug 13705).Aurelien Jarno2012-04-081-2/+2
|
* Disable use of FMA instructions in branredAndreas Schwab2012-03-211-0/+4
|
* Fix pow of zero and infinity to large powers.Joseph Myers2012-03-211-1/+1
|
* Fix atan2 spurious exceptions (bug 11451).Joseph Myers2012-03-191-1/+8
|
* Create and use libc_feupdateenv_test.Richard Henderson2012-03-192-15/+22
| | | | | We can reduce the number of STMXCSR, and often we can avoid the call to __feraiseexcept.
* Create and use SET_RESTORE_ROUND{,_NOEX,_53BIT}{,F,L}.Richard Henderson2012-03-195-64/+53
|
* Make inline __isnan, __isinf_ns, __finite generic.Richard Henderson2012-03-191-0/+35
| | | | | | For code generation to stay identical on x86_64, this requires that we define the fp word manipulation macros before including the generic header.
* Make lround the alias of llround for wordsize-64H.J. Lu2012-03-162-80/+80
|
* Fix tan, tanl for large inputs.Joseph Myers2012-03-161-2/+2
|
* Implement ldbl-96 sinl / cosl / sincosl (bug 13851).Joseph Myers2012-03-161-7/+10
|
* Implement and use libc_feholdexcept_setround_53bit and libc_feupdateenv_53bitAndreas Jaeger2012-03-141-5/+5
| | | | so that double arithmetic in s_sin is done in 53 bit (without extend i386 double precision)
* Clean up dbl-64 rint, nearbyint.Joseph Myers2012-03-134-84/+6
|
View Lorry Controller Status