 masta
|
80d2ee8ad0
Don't reuse registers on ARM 64 bits shift
|
há 11 anos atrás |
 lacak
|
cdc5e9a73a
fcl-db: tests: Improve TestOldValue; tests also Blobs (which has special storage) and state of TField.OldValue after MergeChangeLog
|
há 11 anos atrás |
 sergei
|
9c1f917e3a
* a_call_ref functionality cannot be implemented efficiently at code generator level, because references need specific preparations at earlier points. Moved this support to tcgcallnode and its x86 descendants, and got rid of all ifdef's around.
|
há 11 anos atrás |
|
sergei
|
0d3f36eebf
- Remove references to global variable 'cg' from methods of tcg and some of its descendants.
|
há 11 anos atrás |
|
sergei
|
7104157a57
* Changed tabstractprocdef.address_size to return the def, which can be used to get other properties like size and alignment. Removes one of i8086 ifdef's.
|
há 11 anos atrás |
|
sergei
|
53556f529d
* gen_alloc_symtable: don't use current_procinfo.procdef, the necessary procdef is passed in 'pd' parameter.
|
há 11 anos atrás |
 marco
|
b7d94992eb
* moved unxsockh for solaris, mantis #25657
|
há 11 anos atrás |
 michael
|
7afd5f6afe
* Fixed bug ID #25091
|
há 11 anos atrás |
|
sergei
|
949907f1f3
* Don't emit "Temps allocated between..." comment in assembler listing if procedure has no temps.
|
há 11 anos atrás |
|
sergei
|
1626667374
* Object helper routines review/improvement for alignment-sensitive targets:
|
há 11 anos atrás |
 nickysn
|
f4f35ad47b
* changed ExceptTryLevel from longint to nativeint (because i8086)
|
há 11 anos atrás |
|
sergei
|
e0082262cc
* Don't generate VMT checking code twice (gen_load_vmt_register generates it already).
|
há 11 anos atrás |
|
sergei
|
a5d5ce8eb2
* TP-style object destructors do not need actual VMT in hidden parameter, a constant flag indicating need to free memory is sufficient.
|
há 11 anos atrás |
 florian
|
be17962748
* delete test file immediately after they were used, external tools might get into trouble trying deleting them due to the uncommon names
|
há 11 anos atrás |
|
marco
|
e7639e582c
* try fix #25657
|
há 11 anos atrás |
 Jonas Maebe
|
868b4f50e9
* use spilling_operation_type() instead of a big table to determine whether
|
há 11 anos atrás |
|
Jonas Maebe
|
de0ced1522
* A_INSLWI, A_INSLWI_, A_INSRWI, A_INSRWI_ also modify operand 0
|
há 11 anos atrás |
|
Jonas Maebe
|
6a4afd3a33
* changed uses of "exit" in constructors to if/then constructs, to
|
há 11 anos atrás |
|
Jonas Maebe
|
4e9c54278e
* the function result of constructors always is self, always is valid and is
|
há 11 anos atrás |
|
florian
|
e6b68d19c8
* reordered decision how arguments for inlining are handled
|
há 11 anos atrás |
 Jeppe Johansen
|
07b2982e77
Don't do ARM FoldShiftLdrStr peephole optimization if there's an offset in the reference.
|
há 11 anos atrás |
|
florian
|
2a7dcf6c7b
* formatting
|
há 11 anos atrás |
|
florian
|
e3f9982ee1
* make_not_regable is a matter of code generation so do it in pass_1
|
há 11 anos atrás |
|
florian
|
fbcfb63826
+ implement tcasenode.pass_typecheck
|
há 11 anos atrás |
|
florian
|
d71bfe4c6f
* var parameters can be put in registers if the callee is inlined
|
há 11 anos atrás |
|
florian
|
cd6f4d4469
* fix tsubscriptnode to handle left.location being an fpu or mm register as well.
|
há 11 anos atrás |
 Tomas Hajny
|
77967432ef
* always use forward slashes in path to fpmake similarly to paths to other tools in Makefiles
|
há 11 anos atrás |
|
masta
|
81517fdf70
Rerun peephole optimizers on the whole block
|
há 11 anos atrás |
|
marco
|
4fa8998541
* add SMsgDlgClose, Patch by stocki Mantis #25616
|
há 11 anos atrás |
|
masta
|
3f8549365a
Don't schedule LDRD on ARM
|
há 11 anos atrás |
