 J. Gareth "Curious Kit" Moreton
|
9077954762
* "no_exit_statement_in_block" replaced with "not has_child_of_type(..., [exitn])"
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
59ff8a0f65
* Node-stripping optimisation in "TBlockNode.simplify" now uses "has_child_of_type" to determine stop point
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
ac0e641ce7
* Stripped down and refactored TAsmNode XML node dumps
|
il y a 1 an |
 florian
|
1e3865a187
* patch by rasberryrabbit to resolve #40761
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
e7145f5f7c
* Blocks of statements are now pruned (within reason) if a raise, exit, break, continue or goto node is found
|
il y a 1 an |
 Michaël Van Canneyt
|
1e9e533e6c
* Replace object with record where possible
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
edc1c79cbb
* Bug fix where copied block nodes didn't inherit the original's flags
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
ac3425a14d
* Only perform the block flattening if "forinline" is true or under -O2 and above
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
63e8633262
* New block node optimisation that attempts to
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
5a9d4d44fa
* Calling "internalstatements" now sets the "bnf_strippable" flag on the result
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
060a7185ab
* Framework for more complex block node simplification
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
5605566e42
* Added field to identify blocknodes that wrap entire subroutines
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
2b7df4237b
* nf_pass1_done, nf_error, nf_processing and nf_do_not_execute
|
il y a 1 an |
|
florian
|
4103b3d180
* formatting
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
0e5da37533
* Flags specific to TAsmNode have been moved to their own field
|
il y a 1 an |
|
J. Gareth "Curious Kit" Moreton
|
7854152304
* Adjustment to TBLockNode.pass_1 to actually strip nothing nodes
|
il y a 2 ans |
 Sven/Sarah Barth
|
e050a019a2
* fix #29859: ensure that it's correctly passed down into generate_specialization whether a unit name was provided for the specialization and if so search the generic only in that unit
|
il y a 3 ans |
 Jonas Maebe
|
3e98429418
Revert " * patch by J. Gareth Moreton: Nothing (NOP) node optimisation, resolves #38194"
|
il y a 4 ans |
|
florian
|
c2ff85ac5b
* patch by J. Gareth Moreton: Nothing (NOP) node optimisation, resolves #38194
|
il y a 4 ans |
 yury
|
64c586b86d
* Removed/ifdefed lots of unused variables.
|
il y a 4 ans |
|
florian
|
85afaf5289
* patch by J. Gareth Moreton: Unitialized result variable fix on non-x86 debug builds, resolves #38371
|
il y a 4 ans |
|
florian
|
089e13396a
* ensure that changed compiler options do not cause values
|
il y a 4 ans |
|
florian
|
637976e83f
* patch by Marģers to unify internal error numbers, resolves #37888
|
il y a 5 ans |
|
Jonas Maebe
|
a88288bab3
- reverted r46457, accidentally committed (breaks -Oooptcse)
|
il y a 5 ans |
|
Jonas Maebe
|
f26735fc77
* execute the tempinitcode whent the tempcreate node gets processed rather
|
il y a 5 ans |
|
florian
|
2500f8432f
* ti_const must be part of tempinfostoreflags as well
|
il y a 5 ans |
|
florian
|
62c417c8bf
* temporarily fix #37034 by disabling the problematic optimization until the real cause is fixed
|
il y a 5 ans |
|
florian
|
84a1f614d9
* fix several double ;
|
il y a 5 ans |
|
florian
|
de805fddcd
* properly force the result type after r45217
|
il y a 5 ans |
|
florian
|
e3870079b2
* remove tempref mess for simple inlined functions
|
il y a 5 ans |
