 florian
|
7f6aa98553
* restored changes made in r24864 which I accidently overwrote with r24871
|
há 12 anos atrás |
|
florian
|
980563d576
* several small errors fixed, float128<->float64 conversion works now
|
há 12 anos atrás |
|
florian
|
92477a14b8
o patch by Max Nazholov:
|
há 12 anos atrás |
 pierre
|
aae318403d
Fix a int64<->single conversion failure exposed in test/units/softfpu/sfttst.pp
|
há 12 anos atrás |
|
florian
|
f0487f8fde
* remove packed used for nan records in the softfpu code, it served no purpose
|
há 12 anos atrás |
|
florian
|
b01e6e1671
* remove duplicate types after r20603, resolves #21548
|
há 13 anos atrás |
|
pierre
|
702db31dc6
Fix syntax error in record
|
há 13 anos atrás |
|
florian
|
f0f64a573b
* don't pack softfpu helper records and added dummy for proper alignment, generates much better code on CPUs like arm
|
há 13 anos atrás |
|
pierre
|
f4a9b3d73b
+ Allow compilation as an external unit
|
há 13 anos atrás |
|
florian
|
41636711e4
* fix to align float64 correctly
|
há 14 anos atrás |
|
pierre
|
144d4c41f2
* Define FPC_SYSTEM_HAS_FLOAT32/64 in softfpu unit
|
há 14 anos atrás |
 tom_at_work
|
28e5db2df8
+ Add qword_to_float32/64 functions and fixed int64_to_float64 function
|
há 14 anos atrás |
|
pierre
|
cee190b1f8
* shift128right last parameter should also by var type
|
há 15 anos atrás |
|
florian
|
2a5c3a3c14
* fixed copy&paste typo
|
há 17 anos atrás |
|
florian
|
4ac1deb50b
* moved float_rounding_mode to systemh.inc; renamed to softfloat_rounding_mode
|
há 17 anos atrás |
 peter
|
911f926274
* disabled float128
|
há 18 anos atrás |
|
florian
|
e3085015f7
* continued to work on float128 stuff
|
há 18 anos atrás |
|
florian
|
3afb948ca4
+ start of a float128 unit, was still lying around, so it won't get lost
|
há 18 anos atrás |
|
florian
|
370f05c516
* soft float128 code compiles in theory (ies the compiler)
|
há 18 anos atrás |
|
florian
|
334b9b4ee1
* formatting
|
há 18 anos atrás |
 yury
|
4da2e1a04f
* fixed conversion zero int64 to double in softfloat mode.
|
há 18 anos atrás |
|
florian
|
041ff166cc
* fixed arm compilation
|
há 18 anos atrás |
|
florian
|
26ffc101da
* fixed softfloat trunc
|
há 18 anos atrás |
|
florian
|
b3a1868ff0
* correct masking of exceptions in genmath code
|
há 19 anos atrás |
|
florian
|
1e75956435
* continued to work on float128 translation, tedious work
|
há 19 anos atrás |
|
florian
|
7365009a64
* continued on float128 softfpu code
|
há 19 anos atrás |
 Károly Balogh
|
f4a21e599a
+ fixed this to compile at least on bigendian
|
há 19 anos atrás |
|
florian
|
df5a5cf75e
* started to fix compilation
|
há 19 anos atrás |
|
florian
|
454120a504
+ raw conversion of float80 and float128 support commited
|
há 19 anos atrás |
|
florian
|
4cbb67aa00
* some fpu emulation code from arm to generic code generator moved
|
há 19 anos atrás |
