summaryrefslogtreecommitdiff
path: root/src (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* ARM: Add basic loop and branch instructions.Mike Pall2011-03-301-5/+134
|
* ARM: Add basic unary bytecode instructions.Mike Pall2011-03-291-10/+97
|
* ARM: Add support to call Lua functions and return from them.Mike Pall2011-03-291-3/+107
|
* ARM: Add support to call C functions.Mike Pall2011-03-291-1/+29
|
* ARM: Add entry and exit points into VM. Add type check macros.Mike Pall2011-03-291-10/+141
|
* ARM: Add frame unwind info for assembler part.Mike Pall2011-03-291-1/+37
|
* ARM: Add instruction/call decode + dispatch macros.Mike Pall2011-03-291-0/+79
|
* ARM: Add register assignments, type definitions and stack layout.Mike Pall2011-03-292-7/+70
|
* Avoid compiler warning.Mike Pall2011-03-281-0/+2
|
* Clean up DynASM glue macros. Thanks to Josh Haberman.Mike Pall2011-03-281-14/+0
|
* ARM: Add skeleton for ARM interpreter.Mike Pall2011-03-261-0/+866
|
* ARM: Add support for ARM relocations to buildvm.Mike Pall2011-03-261-2/+12
|
* DUALNUM: Fix narrowing of unary minus.Mike Pall2011-03-191-7/+8
|
* Fix data-flow analysis for BC_ITERC.Mike Pall2011-03-191-1/+2
|
* x64: Use external unwinding for lua_yield().Mike Pall2011-03-182-11/+18
|
* FFI: Fix marking of ffi.gc() finalizer table.Mike Pall2011-03-171-1/+1
|
* x64: Workaround for libgcc unwind bug (still present in RHEL 5.5).Mike Pall2011-03-176-3303/+3331
|
* DUALNUM: Narrow result of math.floor() and math.ceil().Mike Pall2011-03-151-3/+11
|
* DUALNUM: Narrow unary minus.Mike Pall2011-03-154-3/+30
|
* FFI: Force snapshot after C call to preserve semantics.Mike Pall2011-03-151-0/+1
|
* DUALNUM: Add integer variant of MIN/MAX.Mike Pall2011-03-116-7/+64
|
* Get rid of the remaining silly cast macros from Lua.Mike Pall2011-03-1017-65/+58
|
* DUALNUM: Handle integer type in JIT compiler.Mike Pall2011-03-1016-276/+484
|
* FFI: Fix and optimize recording of cdata[cdata].Mike Pall2011-03-102-4/+41
|
* x86/x64: Fix code generation for fused IR_MUL/IR_MULOV.Mike Pall2011-03-101-1/+1
|
* FFI: Fix x86 code generation for ffi.string().Mike Pall2011-03-101-1/+1
|
* Fix handling of err_loc() within cpcall frames.Mike Pall2011-03-091-5/+8
|
* DUALNUM: Add extra assertions to interpreter.Mike Pall2011-03-094-1671/+1738
|
* Fix code generation for bit.bnot().Mike Pall2011-03-071-1/+1
|
* DUALNUM: Make overflow guards weak. Add IR_USE and IR_MULOV.Mike Pall2011-03-074-4/+40
|
* DUALNUM: Fix x87 code for BC_ISEQN/BC_ISNEN.Mike Pall2011-03-074-1257/+1253
|
* Suppress recording of retried fast functions.Mike Pall2011-03-073-2/+9
|
* Fix narrowing of POW.Mike Pall2011-03-031-8/+12
|
* Fix code generation for Intel Atom in x64 mode.Mike Pall2011-03-031-1/+1
|
* FFI: Resolve ld script redirection in ffi.load().Mike Pall2011-03-011-1/+32
|
* FFI: Fix compiled ffi.string() semantics.Mike Pall2011-02-285-5/+8
|
* FFI: Add ffi.gc() function for finalization of cdata objects.Mike Pall2011-02-288-57/+163
|
* DUALNUM: Handle integer type in x86/x64 interpreter and libraries.Mike Pall2011-02-2716-3764/+6439
|
* DUALNUM: Handle integer type in FFI.Mike Pall2011-02-275-17/+42
|
* x64: Improve accuracy of x^-k.Mike Pall2011-02-254-1794/+1791
|
* Fix table.maxn().Mike Pall2011-02-231-1/+1
|
* FFI: Record calls to functions with void results.Mike Pall2011-02-231-2/+8
|
* Eliminate dead slots in snapshots using bytecode data-flow analysis.Mike Pall2011-02-224-21/+133
|
* From Lua 5.2: Support load(string).Mike Pall2011-02-201-1/+4
|
* From Lua 5.2: fp:read("*L").Mike Pall2011-02-201-7/+7
|
* From Lua 5.2: table.unpack(). Needs -DLUAJIT_ENABLE_LUA52COMPAT.Mike Pall2011-02-201-0/+4
|
* From Lua 5.2: Add package.searchpath().Mike Pall2011-02-201-10/+29
| | | | Thanks to F. Perrad.
* From Lua 5.2: Empty statement. Needs -DLUAJIT_ENABLE_LUA52COMPAT.Mike Pall2011-02-201-0/+5
|
* From Lua 5.2: coroutine.running(). Needs -DLUAJIT_ENABLE_LUA52COMPAT.Mike Pall2011-02-201-1/+7
|
* FFI: Fix recording of userdata conversions.Mike Pall2011-02-201-1/+1
|