Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | Removed commented code | Siddhesh Poyarekar | 2013-03-26 | 11 | -85/+0 | |
| | ||||||
* | Make mantissa type of mp_no configurable | Siddhesh Poyarekar | 2013-03-26 | 3 | -69/+101 | |
| | | | | | | | | | | The mantissa of mp_no is intended to take only integral values. This is a relatively good choice for powerpc due to its 4 fpus, but not for other architectures, which suffer due to this choice. This change makes the default mantissa a long integer and allows powerpc to override it. Additionally, some operations have been optimized for integer manipulation, resulting in a significant improvement in performance. | |||||
* | BZ#13889: expl (709.75) wrongly overflows for ldbl-128ibm | Adhemerval Zanella | 2013-03-22 | 2 | -6/+24 | |
| | | | | | | The patch increase the high value to check if expl overflows. Current high mark value is not really correct, the algorithm accepts high values. It also adds a correct wrapper function to check for overflow and underflow. | |||||
* | Fix Bessel function spurious overflows for ldbl-128 / ldbl-128ibm (bug 15285). | Joseph Myers | 2013-03-21 | 2 | -20/+34 | |
| | ||||||
* | Replace 8388608.0 with HALFRAD in mp code | Siddhesh Poyarekar | 2013-03-21 | 2 | -3/+3 | |
| | | | | Minor cleanup | |||||
* | Fix y1l spurious overflows for ldbl-96 (bug 15283). | Joseph Myers | 2013-03-16 | 1 | -1/+1 | |
| | ||||||
* | Fix spurious underflow exceptions for Bessel functions for ldbl-128 / ↵ | Joseph Myers | 2013-03-16 | 2 | -58/+79 | |
| | | | | ldbl-128ibm (bug 14155). | |||||
* | Better distinguish between NaN/qNaN/sNaN. | Thomas Schwinge | 2013-03-15 | 1 | -5/+5 | |
| | ||||||
* | Revert configurable mantissa patch | Siddhesh Poyarekar | 2013-03-15 | 3 | -92/+60 | |
| | | | | | | Reverts d22ca8cdfb98001d03772ef264b244930d439b3f since it is severely broken on 32-bit. | |||||
* | Make mantissa type configurable | Siddhesh Poyarekar | 2013-03-15 | 3 | -60/+92 | |
| | | | | | This allows the default mantissa to be integral, with powerpc overriding it to take advantage of its FPUs. | |||||
* | Fix spurious underflow exceptions for Bessel functions for double (bug 14155). | Joseph Myers | 2013-03-14 | 2 | -4/+8 | |
| | ||||||
* | PowerPC: unify math_ldbl.h implementations | Adhemerval Zanella | 2013-03-08 | 1 | -2/+8 | |
| | | | | | This patch removes redudant definition from PowerPC specific math_ldbl, using the definitions from ieee754 math_ldbl.h. | |||||
* | Merge powerpc slowexp.c into generic code | Siddhesh Poyarekar | 2013-03-07 | 1 | -3/+9 | |
| | ||||||
* | Merge powerpc slowpow.c into generic code | Siddhesh Poyarekar | 2013-03-07 | 1 | -0/+17 | |
| | ||||||
* | Use generic mpa.c code for everything except __mul and __sqr | Siddhesh Poyarekar | 2013-03-07 | 1 | -0/+4 | |
| | ||||||
* | BZ #15055: Use __ieee754_sqrl in acoshl for lbdl-128ibm | Adhemerval Zanella | 2013-03-04 | 1 | -1/+1 | |
| | ||||||
* | Format mpsqrt.c | Siddhesh Poyarekar | 2013-02-27 | 1 | -34/+44 | |
| | ||||||
* | Format mpatan2.c | Siddhesh Poyarekar | 2013-02-27 | 1 | -23/+22 | |
| | ||||||
* | Format mpatan.c | Siddhesh Poyarekar | 2013-02-27 | 1 | -46/+61 | |
| | ||||||
* | Format mptan.c | Siddhesh Poyarekar | 2013-02-27 | 1 | -13/+15 | |
| | ||||||
* | Format mplog.c | Siddhesh Poyarekar | 2013-02-27 | 1 | -21/+23 | |
| | ||||||
* | Another tweak to the multiplication algorithm | Siddhesh Poyarekar | 2013-02-26 | 1 | -6/+50 | |
| | | | | | Reduce the formula to calculate mantissa so that we reduce the net number of multiplications performed. | |||||
* | Don't duplicate mpone and mptwo | Siddhesh Poyarekar | 2013-02-26 | 1 | -5/+3 | |
| | ||||||
* | Remove commented declarations | Siddhesh Poyarekar | 2013-02-25 | 1 | -14/+0 | |
| | ||||||
* | Use long wherever possible in mpa.c | Siddhesh Poyarekar | 2013-02-25 | 1 | -31/+39 | |
| | | | | | | Using long throughout like powerpc does is beneficial since it reduces the need to switch to 32-bit instructions. It gives a very minor performance improvement. | |||||
* | Format slowexp.c | Siddhesh Poyarekar | 2013-02-25 | 1 | -22/+26 | |
| | ||||||
* | Reformat slowpow.c | Siddhesh Poyarekar | 2013-02-25 | 1 | -31/+48 | |
| | ||||||
* | Remove commented code | Siddhesh Poyarekar | 2013-02-25 | 1 | -6/+0 | |
| | ||||||
* | Fix whitespace differences between generic and powerpc mpa.c | Siddhesh Poyarekar | 2013-02-21 | 1 | -2/+2 | |
| | ||||||
* | Respect the user's namespace in installed header files. | Thomas Schwinge | 2013-02-20 | 1 | -1/+1 | |
| | | | | Fixup for commit c7b275d6b3bceb6b400fa3044d13d1001bc605ca. | |||||
* | bits/nan.h: Change __attribute_used__ to __attribute__ ((unused)). | Thomas Schwinge | 2013-02-20 | 1 | -1/+1 | |
| | ||||||
* | Copy comment about inner loop from powerpc mpa.c to the default one | Siddhesh Poyarekar | 2013-02-20 | 1 | -0/+10 | |
| | ||||||
* | Don't require LIM to determine loop end in __sqr | Siddhesh Poyarekar | 2013-02-16 | 1 | -10/+4 | |
| | ||||||
* | Use intermediate variable to compute exponent in __mul | Siddhesh Poyarekar | 2013-02-16 | 1 | -4/+15 | |
| | ||||||
* | Fix determination of lower precision in __mul | Siddhesh Poyarekar | 2013-02-15 | 1 | -1/+4 | |
| | ||||||
* | Use __sqr instead of __mul wherever possible | Siddhesh Poyarekar | 2013-02-14 | 3 | -4/+4 | |
| | ||||||
* | New __sqr function as a faster special case of __mul | Siddhesh Poyarekar | 2013-02-14 | 3 | -2/+94 | |
| | ||||||
* | Remove unnecessary factorial array | Siddhesh Poyarekar | 2013-02-13 | 1 | -10/+3 | |
| | | | | kf is n! at the end of the loop, so storing the values is unnecessary. | |||||
* | Better exp polynomial | Siddhesh Poyarekar | 2013-02-13 | 1 | -23/+37 | |
| | | | | The lesser the __mul calls, the better it is for performance. | |||||
* | Optimized mp multiplication | Siddhesh Poyarekar | 2013-02-13 | 1 | -8/+48 | |
| | | | | Don't bother multiplying zeroes since that only wastes cycles. | |||||
* | Clean up add_magnitudes and sub_magnitudes | Siddhesh Poyarekar | 2013-02-13 | 1 | -52/+60 | |
| | ||||||
* | Remove lots of inline keywords. | Roland McGrath | 2013-02-07 | 1 | -1/+1 | |
| | ||||||
* | Simplify calculation of 2^-m in __mpexp | Siddhesh Poyarekar | 2013-01-18 | 2 | -29/+34 | |
| | ||||||
* | Remove unnecessary multiplication with RADIXI | Siddhesh Poyarekar | 2013-01-18 | 1 | -3/+3 | |
| | ||||||
* | Fix header comment | Siddhesh Poyarekar | 2013-01-17 | 1 | -2/+0 | |
| | ||||||
* | Consolidate constant defines into mpa.h | Siddhesh Poyarekar | 2013-01-16 | 13 | -120/+26 | |
| | ||||||
* | Fix the value of TWO | Siddhesh Poyarekar | 2013-01-16 | 1 | -1/+1 | |
| | ||||||
* | Use libc_fe*() in 32-bit nearbyint() | David S. Miller | 2013-01-15 | 1 | -4/+4 | |
| | | | | | * sysdeps/ieee754/dbl-64/s_nearbyint.c (__nearbyint): Use libc_feholdexcept and libc_fesetenv. | |||||
* | Minor tweak to mp multiplication | Siddhesh Poyarekar | 2013-01-14 | 1 | -13/+15 | |
| | | | | | Add a local variable to remove extra copies to/from memory in the Z array. | |||||
* | Fix code formatting in mpa.c | Siddhesh Poyarekar | 2013-01-14 | 1 | -228/+463 | |
| | | | | This includes the overridden mpa.c in power4. |