Commit Graph

666 Commits

Author SHA1 Message Date
Roberto Ierusalimschy 84e32ad2eb Added opcodes for arithmetic with K operands
Added opcodes for all seven arithmetic operators with K operands
(that is, operands that are numbers in the array of constants of
the function). They cover the cases of constant float operands
(e.g., 'x + .0.0', 'x^0.5') and large integer operands (e.g.,
'x % 10000').
2018-11-23 12:23:45 -02:00
Roberto Ierusalimschy 35296e1fde Details
comments and other janitorial work.
2018-11-22 13:56:04 -02:00
Roberto Ierusalimschy 7f6f70853c To-be-closed variable in 'for' loop separated from the state
The variable to be closed in a generic 'for' loop now is the
4th value produced in the loop initialization, instead of being
the loop state (the 2nd value produced). That allows a loop to
use a state with a '__toclose' metamethod but do not close it.
(As an example, 'f:lines()' might use the file 'f' as a state
for the loop, but it should not close the file when the loop ends.)
2018-11-07 14:42:05 -02:00
Roberto Ierusalimschy 5e76a4fd31 New macros for arithmetic/bitwise operations in 'luaV_execute'
The repetitive code of the arithmetic and bitwise operators in
the main iterpreter loop was moved to appropriate macros.
(As a detail, the function 'luaV_div' was renamed 'luaV_idiv',
as it does an "integer division" (floor division).
2018-11-05 16:10:42 -02:00
Roberto Ierusalimschy e8c779736f Removed internal cache for closures
The mechanism of "caching the last closure created for a prototype to
try to reuse it the next time a closure for that prototype is created"
was removed. There are several reasons:

- It is hard to find a natural example where this cache has a measurable
impact on performance.

- Programmers already perceive closure creation as something slow,
so they tend to avoid it inside hot paths. (Any case where the cache
could reuse a closure can be rewritten predefining the closure in some
variable and using that variable.)

- The implementation was somewhat complex, due to a bad interaction
with the generational collector. (Typically, new closures are new,
while prototypes are old. So, the cache breaks the invariant that
old objects should not point to new ones.)
2018-11-01 13:21:00 -03:00
Roberto Ierusalimschy 947a372f58 State in generic 'for' acts as a to-be-closed variable
The implicit variable 'state' in a generic 'for' is marked as a
to-be-closed variable, so that the state will be closed as soon
as the loop ends, no matter how.

Taking advantage of this new facility, the call 'io.lines(filename)'
now returns the open file as a second result. Therefore,
an iteraction like 'for l in io.lines(name)...' will close the
file even when the loop ends with a break or an error.
2018-10-31 14:54:45 -03:00
Roberto Ierusalimschy e073cbc2e5 Better error messages for invalid operands in numeric 'for'
"Better" and similar to error messages for invalid function arguments.
  *old message: 'for' limit must be a number
  *new message: bad 'for' limit (number expected, got table)
2018-10-30 15:46:56 -03:00
Roberto Ierusalimschy 6e9b719694 More uniformity in code generation for 'for' loops
Added new instruction 'OP_TFORPREP' to prepare a generic for loop.
Currently it is equivalent to a jump (but with a format 'iABx',
similar to other for-loop preparing instructions), but soon it will
be the place to create upvalues for closing loop states.
2018-10-26 10:38:50 -03:00
Roberto Ierusalimschy 41c800b352 Closing methods should not interfere with returning values
A closing method cannot be called in its own stack slot, as there may
be returning values in the stack after that slot, and the call would
corrupt those values. Instead, the closing method must be copied to the
top of the stack to be called.

Moreover, even when a function returns no value, its return istruction
still has to have its position (which will set the stack top) after
the local variables, otherwise a closing method might corrupt another
not-yet-called closing method.
2018-10-25 12:50:20 -03:00
Roberto Ierusalimschy 3c7dc52909 Handling of memory errors when creating to-be-closed upvalues 2018-10-18 16:15:09 -03:00
Roberto Ierusalimschy bd96330d03 First "complete" implementation of to-be-closed variables
Still missing:
- handling of memory errors when creating upvalue (must run closing
method all the same)
- interaction with coroutines
2018-10-17 10:44:42 -03:00
Roberto Ierusalimschy 4cd1f4aac0 Towards "to closed" local variables
Start of the implementation of "scoped variables" or "to be closed"
variables, local variables whose '__close' (or themselves) are called
when they go out of scope. This commit implements the syntax, the
opcode, and the creation of the corresponding upvalue, but it still
does not call the finalizations when the variable goes out of scope
(the most important part).

Currently, the syntax is 'local scoped name = exp', but that will
probably change.
2018-10-08 10:42:07 -03:00
Roberto Ierusalimschy 5382a22e0e Corrections in the implementation of '%' for floats.
The multiplication (m*b) used to test whether 'm' is non-zero and
'm' and 'b' have different signs can underflow for very small numbers,
giving a wrong result. The use of explicit comparisons solves this
problem. This commit also adds several new tests for '%' (both for
floats and for integers) to exercise more corner cases, such as
very large and very small values.
2018-08-28 12:36:58 -03:00
Roberto Ierusalimschy 8c8a91f2ef Deprecated the emulation of '__le' using '__lt'
As hinted in the manual for Lua 5.3, the emulation of the metamethod
for '__le' using '__le' has been deprecated. It is slow, complicates
the logic, and it is easy to avoid this emulation by defining a proper
'__le' function.

Moreover, often this emulation was used wrongly, with a programmer
assuming that an order is total when it is not (e.g., NaN in
floating-point numbers).
2018-08-24 10:17:54 -03:00
Roberto Ierusalimschy 3dcd04ad61 details
Minor optimizations in 'lvm.c'. (Not exactly optimizations, but more
chances for optimizations.)
2018-08-17 15:53:39 -03:00
Roberto Ierusalimschy 06e08c6d05 Fixed bug in OP_IDIVI
Opocode was using 'luai_numdiv' (float division) instead of
'luai_numidiv' (integer division).
2018-07-09 12:41:24 -03:00
Roberto Ierusalimschy a314409dba in generational mode, an emergency collection can turn any object black
during any memory allocation +
'luaT_getvarargs' may reallocate the stack, and therefore the top must
be correct.
2018-06-18 14:58:21 -03:00
Roberto Ierusalimschy 6e600695f8 field 'sizearray' in struct 'Table' changed to 'alimit', which can
be used as a hint for '#t'
2018-06-15 11:14:20 -03:00
Roberto Ierusalimschy fb8fa66136 no more 'luaH_emptyobject' and comparisons of addresses of global variables
(instead, use a different kind of nil to signal the fake entry returned
when a key is not found in a table)
2018-06-01 13:51:34 -03:00
Roberto Ierusalimschy 34aa0c5bd7 new macros 'likely'/'unlikely' with hints for jump predictions
(used only in errors for now)
2018-05-30 11:25:52 -03:00
Roberto Ierusalimschy 02ed0b2c30 in 'luaD_poscall', there is no need to compute 'firstResult' when 'nres==0' 2018-05-22 09:02:36 -03:00
Roberto Ierusalimschy e64e20ac81 minimizing the code ran by 'vmfetch' + no more 'vra'
(the code is simpler without 'vra' and conversion is a no-op)
2018-05-02 15:17:59 -03:00
Roberto Ierusalimschy 03c6a05ec8 no more nil-in-table 2018-04-04 11:23:41 -03:00
Roberto Ierusalimschy 3d0b5edfe4 using unsigned comparison in 'l_intfitsf' (avoids one comparison) 2018-04-02 14:52:07 -03:00
Roberto Ierusalimschy 7b0b6b3b39 cannot use 'defined' inside a macro +
call to 'luaT_keydef' must be protected
2018-03-16 11:21:20 -03:00
Roberto Ierusalimschy 4a1612ff9b new experimental syntax using reserved word 'undef' 2018-03-07 12:55:38 -03:00
Roberto Ierusalimschy 62a392ff46 using jump tables when available 2018-03-02 15:59:19 -03:00
Roberto Ierusalimschy ef8263f81f better names for macros for tags and types.
rttype -> rawtt; ttyperaw -> withvariant; ttype -> ttypetag;
tnov -> ttype
2018-02-26 11:16:05 -03:00
Roberto Ierusalimschy 9243c414d9 first version of empty entries in tables
(so that, in the future, tables can contain regular nil entries)
2018-02-23 10:16:18 -03:00
Roberto Ierusalimschy c72ac048b9 conditional jumps "deunified"
(if a jump table is used, the unification may harm jump prediction.)
2018-02-21 16:43:44 -03:00
Roberto Ierusalimschy 212095a601 new opcodes OP_GTI/OP_GEI 2018-02-21 12:49:32 -03:00
Roberto Ierusalimschy 06865aa01d simpler implementation for 'LTintfloat'/'LEintfloat'
+ 'LTfloatint'/'LEfloatint'
2018-02-21 10:47:03 -03:00
Roberto Ierusalimschy 465b474899 small reorganization of 'luaV_flttointeger'/'luaV_tointeger' 2018-02-21 09:54:26 -03:00
Roberto Ierusalimschy 1afd5a152d more generic way to handle 'gclist' 2018-02-19 17:06:56 -03:00
Roberto Ierusalimschy 49dae52d08 correct way to check stack space for vararg functions 2018-02-17 17:20:00 -02:00
Roberto Ierusalimschy 0682fe8169 some simplifications/optimizations in returns from Lua functions 2018-02-15 13:34:29 -02:00
Roberto Ierusalimschy b1379936cf vararg back to '...' (but with another implementation)
new implementation should have zero overhead for non-vararg functions
2018-02-09 13:16:06 -02:00
Roberto Ierusalimschy 318a9a5859 new opcode 'PREPVARARG'
(avoids test for vararg function in all function calls)
2018-02-07 13:18:04 -02:00
Roberto Ierusalimschy 51280ef2ad call hooks for Lua functions called by 'luaV_execute' 2018-02-06 17:16:56 -02:00
Roberto Ierusalimschy dc0ab1e8ca warnings in VS (implicit casts from ptrdiff_t to int) 2018-01-29 14:21:35 -02:00
Roberto Ierusalimschy 5bd8d388de OP_CONCAT does not move its result (to simplify its execution) 2018-01-27 14:56:33 -02:00
Roberto Ierusalimschy d2fb34ac88 'OP_TAILCALL' calling C functions finishes the call and returns
(instead of waiting for following 'OP_RETURN')
2018-01-14 15:27:50 -02:00
Roberto Ierusalimschy ab0a851db4 'luaD_tryfuncTM' can ensure it does not change the stack 2018-01-10 17:19:27 -02:00
Roberto Ierusalimschy ad960095bf avoid jumping into a variable scope (C++ does not allow that) 2018-01-09 12:23:40 -02:00
Roberto Ierusalimschy a9295a2b8e typos in comments 2017-12-30 18:46:18 -02:00
Roberto Ierusalimschy cf7eff45f3 keep control of stack top in Lua functions concentrated in 'luaV_execute' 2017-12-28 13:42:57 -02:00
Roberto Ierusalimschy 4676f6599e new macros 'isOT'/'isIT'
(plus exchanged parameters of OP_VARARG to make it similar to other
'isOT' instructions)
2017-12-22 12:16:46 -02:00
Roberto Ierusalimschy 1d5b885437 when running Lua code, there is no need to keep 'L->top' "correct";
set it only when needed.
2017-12-20 12:58:05 -02:00
Roberto Ierusalimschy 3153a41e33 no need to save 'pc' in case of allocation errors
(allocation errors do not call message handlers)
2017-12-19 14:18:04 -02:00
Roberto Ierusalimschy d388c165ef new opcodes 'FORLOOP1'/'FORPREP1' for "basic for" (integer variable
with increment of 1)
2017-12-18 15:53:50 -02:00