narayana
/
ravi
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

doc update

Browse Source
Dibyendu Majumdar 9 years ago
parent 1292bad982
commit 387d0683f7
1 changed files with 14 additions and 14 deletions
Show Stats Download Patch File Download Diff File

28
readthedocs/ravi-jit-libgccjit.rst
Unescape View File

@ -1,5 +1,5 @@
========================================
JIT Compilation for Ravi using libgccjit
JIT Compilation for Ravi using ``libgccjit``
========================================
Introduction
@ -16,9 +16,9 @@ Why another JIT engine?
-----------------------
Well partly as I feel I have a moral obligation to support gcc, given it has been instrumental in bringing about the OpenSource / Free Software ecosystem.
Secondly I am always looking for alternatives that will let me reduce the footprint of Ravi. The libgccjit is offered as a shared library - this is a great thing. I hate to have to statically link LLVM.
Secondly I am always looking for alternatives that will let me reduce the footprint of Ravi. The ``libgccjit`` is offered as a shared library - this is a great thing. I hate to have to statically link LLVM.
LLVM implementation and libgccjit implementation will both be kept in sync so that user can choose either option.
LLVM implementation and ``libgccjit`` implementation will both be kept in sync so that user can choose either option.
Building GCC
------------
@ -40,12 +40,12 @@ I built gcc 5.1 from source as follows.
Current Status
--------------
Work on this started only recently (8 June 2015) so not much to show yet. But expectation is that there will be a working implementation by end June - the strategy is to port the existing LLVM implementation to equivalent libgccjit implementation.
Work on this started only recently (8 June 2015) so not much to show yet. But expectation is that there will be a working implementation by end June - the strategy is to port the existing LLVM implementation to equivalent ``libgccjit`` implementation.
Building Ravi with libgccjit on Linux
Building Ravi with ``libgccjit`` on Linux
-------------------------------------
.. warning:: The current official distribution of ``libgccjit`` does not work for Ravi due to the fact that unreachable blocks are treated as errors - this prevents Ravi code from being compiled. I have a personal copy where I have disabled this validation - I am hoping that the next release of ``libgccjit`` will make this an optional validation.
.. warning:: The current official distribution of ``libgccjit`` does not work for Ravi due to the fact that unreachable blocks are treated as errors - this prevents Ravi code from being compiled. There is also `a bug <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66700>`_ I have a personal copy where I have disabled this validation - I am hoping that the next release of ``libgccjit`` will make this an optional validation.
.. warning:: Note that right now the Ravi's ``libgccjit`` based JIT implementation is work in progress and has bugs (some Lua tests fail).
@ -69,7 +69,7 @@ Setting up of the Lua types is proving easier in ``libgccjit`` due to the fact t
JIT Status of Lua/Ravi Bytecodes
---------------------------------
Following is the status as of 28 June 2015.
Following is the status as of 30 June 2015.
+-------------------------+----------+--------------------------------------------------+
| name | JITed? | description |
@ -88,17 +88,17 @@ Following is the status as of 28 June 2015.
+-------------------------+----------+--------------------------------------------------+
| OP_GETTABUP | YES | R(A) := UpValue[B][RK(C)] |
+-------------------------+----------+--------------------------------------------------+
| OP_GETTABLE | NO | R(A) := R(B)[RK(C)] |
| OP_GETTABLE | YES | R(A) := R(B)[RK(C)] |
+-------------------------+----------+--------------------------------------------------+
| OP_SETTABUP | NO | UpValue[A][RK(B)] := RK(C) |
+-------------------------+----------+--------------------------------------------------+
| OP_SETUPVAL | NO | UpValue[B] := R(A) |
+-------------------------+----------+--------------------------------------------------+
| OP_SETTABLE | NO | R(A)[RK(B)] := RK(C) |
| OP_SETTABLE | YES | R(A)[RK(B)] := RK(C) |
+-------------------------+----------+--------------------------------------------------+
| OP_NEWTABLE | NO | R(A) := {} (size = B,C) |
| OP_NEWTABLE | YES | R(A) := {} (size = B,C) |
+-------------------------+----------+--------------------------------------------------+
| OP_SELF | NO | R(A+1) := R(B); R(A) := R(B)[RK(C)] |
| OP_SELF | YES | R(A+1) := R(B); R(A) := R(B)[RK(C)] |
+-------------------------+----------+--------------------------------------------------+
| OP_ADD | NO | R(A) := RK(B) + RK(C) |
+-------------------------+----------+--------------------------------------------------+
@ -128,9 +128,9 @@ Following is the status as of 28 June 2015.
+-------------------------+----------+--------------------------------------------------+
| OP_BNOT | NO | R(A) := ~R(B) |
+-------------------------+----------+--------------------------------------------------+
| OP_NOT | NO | R(A) := not R(B) |
| OP_NOT | YES | R(A) := not R(B) |
+-------------------------+----------+--------------------------------------------------+
| OP_LEN | NO | R(A) := length of R(B) |
| OP_LEN | YES | R(A) := length of R(B) |
+-------------------------+----------+--------------------------------------------------+
| OP_CONCAT | YES | R(A) := R(B).. ... ..R(C) |
+-------------------------+----------+--------------------------------------------------+
@ -162,7 +162,7 @@ Following is the status as of 28 June 2015.
+-------------------------+----------+--------------------------------------------------+
| OP_TFORLOOP | NO | if R(A+1) ~= nil then { R(A)=R(A+1); pc += sBx } |
+-------------------------+----------+--------------------------------------------------+
| OP_SETLIST | NO | R(A)[(C-1)*FPF+i] := R(A+i), 1 <= i <= B |
| OP_SETLIST | YES | R(A)[(C-1)*FPF+i] := R(A+i), 1 <= i <= B |
+-------------------------+----------+--------------------------------------------------+
| OP_CLOSURE | YES | R(A) := closure(KPROTO[Bx]) |
+-------------------------+----------+--------------------------------------------------+

Write Preview
Loading…
Cancel
Save