 pierre
|
d1e091d5a3
Use .byte substitutes for ldmxcsr and stmxcsr instructions
|
8 years ago |
 sergei
|
17256e8387
* fpc_exp_real: replaced pushing immediate values on stack with Darwin-style PIC. This change increases performance on Intel E7200 almost twice. On AMD CPUs performance increase is not that large, but still noticeable.
|
9 years ago |
|
sergei
|
fc5f45f65c
* sqr(real) and sqrt(real): remove typeconv node inserted by initial call processing (see explanation in comments), allowing these functions to be evaluated using precision of argument. In particular, sqrt(single) and sqrt(double) now emit 'sqrtss' and 'sqrtsd' instructions on x86 targets with -Cfsse3. Non-x86 targets already have the necessary support in code generators.
|
11 years ago |
|
sergei
|
b16c6f8ced
* i386 and x86_64 changes for Delphi compatibility:
|
11 years ago |
|
sergei
|
6810d643c4
* Fixed exp() result for +-Inf input, must be +Inf or 0, not NaN. This is mathematically correct and consistent with generic implementation of this function.
|
11 years ago |
|
sergei
|
57c762c8d2
+ Inserted license header for AMath routines.
|
11 years ago |
|
sergei
|
bd58adfcc9
+ Added credits for functions reused from AMath/DAMath libraries.
|
11 years ago |
|
sergei
|
217bac7a0b
- Removed fpc_pi_real compilerproc, it is no longer used. At least two releases (2.6.2 and 2.6.4) emit Pi directly as a constant.
|
11 years ago |
|
sergei
|
2981f73aaa
- Removed redundant functions "power", they were neither used nor accessible through interface. The actual function "power" is located in Math unit.
|
11 years ago |
|
sergei
|
d251c131a5
* i386: replaced fpc_exp_real with a port from AMath library. It has better accuracy and is faster on CPUs that suffer penalties from changing x87 control word.
|
11 years ago |
 florian
|
558cf3ef87
* reset FPU properly after an exception, resolves #12214
|
17 years ago |
 Jonas Maebe
|
c35e0bf712
- removed fclex at the end of fpc_exp_real because it makes that routine 10+
|
17 years ago |
 yury
|
7230661978
* Fixed warnings about EBP based access.
|
18 years ago |
|
Jonas Maebe
|
f51bac256d
* made pic-safe (by simply not accessing global variables from
|
18 years ago |
|
florian
|
1cd41b68f7
* compiler handled round/trunc for x86
|
18 years ago |
|
florian
|
47fac4ee6e
* moved *SSECSR to system unit
|
19 years ago |
 michael
|
3a2eaa94b1
+ Removed INTERNCONSTINTF define
|
20 years ago |
|
michael
|
93ba0409be
+ Removed HASCOMPILERPROC define
|
20 years ago |
 peter
|
4ace790492
* remove $Log
|
20 years ago |
 fpc
|
790a4fe2d3
* log and id tags removed
|
20 years ago |
|
fpc
|
50778076c3
initial import
|
20 years ago |
|
florian
|
f94047b703
* some floating point routines improved
|
20 years ago |
|
florian
|
3e5f68d360
* C. Western: utf-8 reading from resource files
|
20 years ago |
|
peter
|
e417e34496
* truncate log
|
20 years ago |
|
peter
|
4d8460ec2f
* float routines all use internproc and compilerproc helpers
|
21 years ago |
|
peter
|
14eb8f59b4
internconst, internproc and some external declarations moved to interface
|
21 years ago |
|
peter
|
783f8068a5
* add fclex for fpu exceptions to round/trunc
|
21 years ago |
|
Jonas Maebe
|
89d4ca293b
* fix power() for negative base
|
22 years ago |
|
michael
|
8373e40e10
+ Patch from Johannes Berg for bug #2759
|
22 years ago |
|
peter
|
8f8b47194a
* REGCALL define added
|
22 years ago |
