Age | Commit message (Collapse) | Author |
|
The try_end and try_case instructions are implemented the same way
(try_case is translated to try_end by the loader).
We can do better than that. We know that try_case will only be executed
when an exception has been caught. Therefore, we know that x(0) is
the non-value and that x(1) through x(3) need to be shifted down to
x(0) through x(2). There is no need to test x(0) before shifting down.
try_end does not need the register shifting code at all.
|
|
Move out from the head the variables that are only used in the excute
phase.
|
|
|
|
Introduce new macros that can be used for relative jumps and
use them consistently.
Test that everything works by using a non-zero constant JUMP_OFFSET.
The loader subtracts JUMP_OFFSET from loaded labels, and all
instructions that use 'f' operands add it back.
|
|
The C compiler will probably optimize this, but just to be sure...
|
|
It's bad style. Pass the name of the variable as an
extra argument to the macro.
|
|
|
|
|
|
We don't need to pass x(0), x(1), and x(2) because they
can already be found in the register array.
|
|
In a correct Erlang programs, we can expect that:
* A GC test instruction (such as test_heap) is more likely
not to do the GC.
* A BIF is more likely to succeed than to fail.
* A BIF is more likely to fail in a guard than in a body.
* An apply or fun call is likely to succeed.
Annotate conditions accordingly.
|
|
|
|
|
|
Eliminate the need to write pre-processor macros for each instruction.
Instead allow the implementation of instruction to be written in
C directly in the .tab files. Rewrite all existing macros in this
way and remove the %macro directive.
|