 pierre
|
0fa280f4c2
Fix code generated in tarmnotnode.second_boolean
|
%!s(int64=6) %!d(string=hai) anos |
 florian
|
45847da693
* fix building for targets with no vfp support
|
%!s(int64=6) %!d(string=hai) anos |
|
florian
|
85edf1c1eb
* reworked arm vfp capability handling to use fpu_capabilites
|
%!s(int64=6) %!d(string=hai) anos |
|
florian
|
b3ed34592f
+ software handling of exceptions on arm
|
%!s(int64=6) %!d(string=hai) anos |
 Jeppe Johansen
|
cba4aeaca5
Fix division of constants on ARM thumb without long multiplication.
|
%!s(int64=6) %!d(string=hai) anos |
|
florian
|
71e71ad267
* fix currency division on non x86 32 bit targets
|
%!s(int64=7) %!d(string=hai) anos |
|
florian
|
cd41312a8f
* fixes not(<qwordbool>) on arm
|
%!s(int64=7) %!d(string=hai) anos |
 nickysn
|
efc5e339d0
* use an enum instead of integer constants to represent inline numbers
|
%!s(int64=8) %!d(string=hai) anos |
|
florian
|
1c067e96bf
* fix VFPv4 support
|
%!s(int64=9) %!d(string=hai) anos |
 yury
|
432248cbf1
* Removed lot of unused vars.
|
%!s(int64=10) %!d(string=hai) anos |
|
Jeppe Johansen
|
dac294c680
Fix ARMv3/ARMv2A support.
|
%!s(int64=10) %!d(string=hai) anos |
 Jonas Maebe
|
10b2ea3b1b
* use handle_locjump() instead of local inlined version
|
%!s(int64=10) %!d(string=hai) anos |
|
Jeppe Johansen
|
db401f0371
Add missing size postfix to VNEG VFP instruction.
|
%!s(int64=10) %!d(string=hai) anos |
|
Jeppe Johansen
|
9e5979e8be
Implemented UAL syntax support in the ARM assembler reader. Can be toggled with a field for now, but not implemented yet. Still using pre-UAL syntax for now.
|
%!s(int64=10) %!d(string=hai) anos |
 masta
|
0cb1a129b3
{ARM} Implement usage of generic division-by-const optimization
|
%!s(int64=11) %!d(string=hai) anos |
 sergei
|
68b97bee5a
* ARM: Implemented floating-point negation and abs() for softfloat using integer instructions. Fixes webtbs/tw4534.pp, and also yields much faster code than existing implementation.
|
%!s(int64=11) %!d(string=hai) anos |
|
Jonas Maebe
|
4065483a50
* completed thlcgobj.location_force_fpureg(), use it everywhere and removed
|
%!s(int64=11) %!d(string=hai) anos |
|
masta
|
a72a3f4e01
Little cleanup after r26669, no functional changes
|
%!s(int64=11) %!d(string=hai) anos |
|
masta
|
80d2ee8ad0
Don't reuse registers on ARM 64 bits shift
|
%!s(int64=11) %!d(string=hai) anos |
|
Jeppe Johansen
|
d24cbbf9f5
Changed debug information to dwarf for ARM_embedded, and set local minimum alignment to 4.
|
%!s(int64=11) %!d(string=hai) anos |
|
Jonas Maebe
|
b6d279d4aa
* don't transform div-by-power-of-2 into a shift when overflow checking is
|
%!s(int64=11) %!d(string=hai) anos |
|
florian
|
73e6af4864
+ cpu flag CPUARM_HAS_THUMB_IDIV
|
%!s(int64=12) %!d(string=hai) anos |
|
florian
|
d4968e054b
+ arm: tsettings.instructionset
|
%!s(int64=12) %!d(string=hai) anos |
|
florian
|
e514e84c83
* first_moddivint should not change the return type if div/mul/add is used
|
%!s(int64=12) %!d(string=hai) anos |
|
Jeppe Johansen
|
570b40faed
Signed modulus by 2 on ARM with no division is optimized to a series of instructions instead of calling fpc_mod_longint.
|
%!s(int64=12) %!d(string=hai) anos |
|
Jonas Maebe
|
5051453806
+ support for LOC_(C)MMREGISTER in hlcg
|
%!s(int64=12) %!d(string=hai) anos |
|
florian
|
21c154d60a
Merged r22903
|
%!s(int64=12) %!d(string=hai) anos |
|
florian
|
1eeeb309c7
* intial armv6m support, it is not working yet, constant pool insertation and conditional branch fixup is not working yet
|
%!s(int64=12) %!d(string=hai) anos |
|
florian
|
47d43750e4
* remove unused units from uses statements
|
%!s(int64=12) %!d(string=hai) anos |
|
Jeppe Johansen
|
3ee29eb219
Fixed ARMv7-EM code generation and RTL compilation
|
%!s(int64=12) %!d(string=hai) anos |
