 Rika Ichinose
|
2d6294eb26
MovQ + Shr → PExtrW.
|
1 year ago |
 Jonas Maebe
|
0758aa1143
FPU exception mask: generlised system unit interface
|
2 years ago |
 florian
|
652e75f8ec
* use v*csr instructions if they are available
|
3 years ago |
|
florian
|
fbf68af605
* continued to update change information
|
3 years ago |
|
florian
|
d354428f42
* moved fwait
|
4 years ago |
|
florian
|
777d536d9b
* extended routines int/frac/trunc shouldn't mess with precision or exception flags, should fix #39275
|
4 years ago |
|
florian
|
f0d92b74c1
- x86_64: some more fwaits removed
|
4 years ago |
|
florian
|
cebf880df9
* x86-64: fpc_frac_real further improved: memory loads of extendeds are expensive
|
4 years ago |
|
florian
|
fafee8b85b
* x86-64: removed some fwait instructions from fpc_frac_real, they shouldn't be necessary
|
4 years ago |
 pierre
|
40b7af1025
Use pascal code for LLVM fpc_sqr_real, because fmultp instruction is not accepted by llvm
|
4 years ago |
|
pierre
|
e35e5ad349
Add missing CPULLVM versions for fpc_abs_real and fpc_sqr_real in x86_64/math.inc
|
4 years ago |
 svenbarth
|
7990b2e3f3
* apply patch from J. Gareth Moreton to implement Int() for SSE (currently only used on Win64)
|
7 years ago |
|
svenbarth
|
abd893cac4
* fix for Mantis #33635: correctly working SSE2 based Frac() implementation by J. Gareth Moreton
|
7 years ago |
|
svenbarth
|
e3ea46d1cb
* disable SSE optimized Frac() for Win64; it has a much lower supported range than the existing Pascal implementation which uses Int() (in essence Round() and Trunc() should be corrected as well)
|
7 years ago |
|
pierre
|
4026644278
Add explicit registers to fcomip to allow compilation with clang (bug report 33547)
|
7 years ago |
|
svenbarth
|
5f08e47d84
+ add an SSE2 implementation of Frac() (the existing Trunc() and Round() implementations are also SSE2, so nothing special to keep in mind here)
|
8 years ago |
|
svenbarth
|
b58c3ed212
+ add guards for consistency with the code for Extended
|
8 years ago |
 sergei
|
cad30aaee9
* x86_64: Modified fpc_exp_real so it does not use 'sahf' assembler instruction, which may be not present on early 64-bit capable x86 processors. Mantis #29583.
|
9 years ago |
|
Jonas Maebe
|
e4647ac133
- removed "nostackframe" from floating point helpers for llvm, as FPC
|
9 years ago |
|
Jonas Maebe
|
6909523b70
* handle atan, sqrt, ln, sin and cos in the x86 RTL for llvm
|
10 years ago |
|
sergei
|
f456bb3a25
* Re-implement ln(x) also for x87-based x86_64 targets (counterpart of r27367,r27518,r27552,r27553 for i386 target).
|
10 years ago |
|
sergei
|
b16c6f8ced
* i386 and x86_64 changes for Delphi compatibility:
|
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
|
2b1e5f7014
* Mantis #17273: don't generate x87 instructions on win64 target.
|
11 years ago |
|
sergei
|
90d66595c2
* typecheck_real_to_currency: generate inline round() node instead of direct call to fpc_round_real, this allows target-specific processing to take place and possibly emit more efficient code.
|
12 years ago |
|
sergei
|
db045c4871
* Define FPC_ABSMASK_SINGLE and FPC_ABSMASK_DOUBLE without using assembler.
|
12 years ago |
|
sergei
|
ea571ba1a3
* Use variables instead of direct indexing into stack, eliminates warnings at compilation.
|
14 years ago |
|
sergei
|
369725f2c4
* Get8087CW: Make sure that high bits of result are clear.
|
14 years ago |
|
sergei
|
4a3c7f7ef4
* Get8087CW: use fixed stack.
|
14 years ago |
