(which are not supported on all unix targets; feel free to re-enable for
specific unix targets if the test has been verified to compile there)
o also fixed placement of dotest directives: they have to come before any
part of the program code
git-svn-id: trunk@41717 -
so that they take into account inherited calling conventions (mantis #35233)
o don't needlessly calculate the paraloc info when generating a JVM mangled
name
git-svn-id: trunk@41716 -
+ compiler/sysmsym.pas: Do the same for for all tsym constructors.
+ compiler/symtype.pas: Generate internalerror in tcompilerppufile.putderef
if a deref field has index -1, as this means that buildderef was not called
while it should have been called.
+ compiler/symtable.pas: Fix bug report itself by adding an extra local variable
CHANGED to add extra cycles in tstoredsymtable.buildderef_referenced method.
New tests for this bug report: tests/webtbs/tw35139.pp and tests/webtbs/tw35139a.pp
git-svn-id: trunk@41425 -
to the global variable in the read-locked sections could cause
false positives regarding detecting whether nothing got changed
during a write-locked section due to lack of read/write dependencies
git-svn-id: trunk@41165 -
as this might result in some nodes not being executed, like temp. create nodes with init. code, see e.g. issue #34653, resolves#34653
git-svn-id: trunk@40934 -
encountering a method with the correct name that does not have the
"overload" directive (same logic as when looking for a call candidate,
to avoid errors when using a Pascal-level wrapper to call interface
methods, and Delphi-compatible since it always required "overload" for
overloaded methods)
o also catches calling convention mismatches like in webtbs/tw27349
git-svn-id: trunk@40683 -
This patch improves the compiler where "case" statements are concerned, using jump tables more often and creating more efficient machine code in some situations:
* If a case block only contains one branch (not including the else block), the initial range check is removed, since this becomes wasted effort.
* If the else block is empty, the else label is set to the end label - though this doesn't decrease the code size, it takes a bit of strain off the peephole optimizer.
* On -O2 and above, some node analysis is now done on the branch labels. Most of the time this just redirects it to the end
label for empty blocks, but if the block contains a goto statement, it will redirect it to its destination instead,
thus increasing performance by not having multiple jumps (this won't get picked up by the peephole optimiser if the label addresses are in a jump table).
* Some checks now use what I call the 'true count' rather than the 'label count'. The true count includes each
individual value in a range - for example, 0..2 counts as 3. This increases the chance that a jump table will be
utilised in situations where it is more efficient than a linear list.
* For jump tables, if the case block almost covers the entire range (32 entries or fewer from full coverage),
the initial range check is removed and the gaps included in the jump table (pointing to the else label).
git-svn-id: trunk@40676 -
