pierre
|
299799c33c
Add name of package for which dependency is checked to get a more useful error message
|
5 years ago |
Jonas Maebe
|
b5b1dc6dea
* fixed ppc64 triple architecture names
|
5 years ago |
Jonas Maebe
|
86d55055df
* made internalerror unique
|
5 years ago |
Jonas Maebe
|
69129f38e2
* aarch64: correct paraloc^.def for LOC_REFERENCE parameters
|
5 years ago |
florian
|
1d3ed354de
* patch by J. Gareth Moreton: unifies internalerrors, resolves #37471
|
5 years ago |
yury
|
719220e8a4
* For 8/16 bit CPUs and TP mode, disable expansion to a larger integer type for subtract operations on unsigned integers. This is TP7 compatible.
|
5 years ago |
michael
|
ec3eef03f1
* Set path to local version of fcl-passrc, not installed version
|
5 years ago |
michael
|
2faa9e06a1
* Additional options for latest pas2js version features
|
5 years ago |
michael
|
869ae0904c
* Lowercase unit name (bug ID 37481)
|
5 years ago |
pierre
|
ca40956a3a
Regenerated all Makefile's
|
5 years ago |
pierre
|
a2806954ed
Update after commit #46222
|
5 years ago |
pierre
|
a5997e266f
Add powerpc specific call to gcc to get the correct libgcc directory for linux
|
5 years ago |
pierre
|
fdb1cd38ab
Disable limitation of handling of negative shift values, can be enalbed again by setting LIMIT_NEG_SHIFTVALUES macro
|
5 years ago |
svenbarth
|
c17ad509e2
* fix for Mantis #36775: apply patch by J. Gareth Moreton
|
5 years ago |
svenbarth
|
055cfc1771
* added test for Mantis #36014 which is already fixed (also in 3.2.0)
|
5 years ago |
svenbarth
|
8770af84bd
* fix for Mantis #37187: inside generics the constant code in pexpr does not handle all cases and thus current_procinfo needs to be checked as well
|
5 years ago |
Jonas Maebe
|
06a5ec5b5c
* fixed peephole optimizer replace lea by add even if the flags were in use
|
5 years ago |
Jonas Maebe
|
3047ce71bc
* fixed LLVM register allocator after r46199
|
5 years ago |
florian
|
87702d7cda
* cosmetics
|
5 years ago |
florian
|
7281ea7a2c
* floating point division by zero should throw run time error 208, fixed
|
5 years ago |
florian
|
b60ac4c2db
* @PLT is needed for calls to link it properly with pic code
|
5 years ago |
florian
|
4c2c540b29
* fixed resetting of state, broken in r45995
|
5 years ago |
florian
|
893225770c
* updated tfpstate
|
5 years ago |
yury
|
0000682043
* i8086: When a ref points to the 32-bit data, preload the ref base to a temp register to reduce spilling.
|
5 years ago |
yury
|
ddea929923
* For 8 and 16 bit CPUs preload the ref base and index to a separate temp register in the generic implementation of a_load64_reg_ref() and a_load64_ref_reg().
|
5 years ago |
yury
|
e70c898503
* Removed the i8086 specific hack. Since r46199 the register allocator is able to handle the problem code.
|
5 years ago |
yury
|
ff820247c3
* Early during register allocation, spill registers which interfere with all usable real registers.
|
5 years ago |
yury
|
601e05f17f
* Count only usable real registers.
|
5 years ago |
yury
|
0daba07246
* Enhanced the debug output of spilling statistics. Now it includes efficiency of the spilling which is based on
|
5 years ago |
florian
|
eac6675cba
* reset exception flags, resolves #37468 on x86_64
|
5 years ago |