aboutsummaryrefslogtreecommitdiff
path: root/lcode.c (follow)
Commit message (Collapse)AuthorAgeFilesLines
* New instruction format for SETLIST/NEWTABLERoberto Ierusalimschy2024-06-281-25/+34
| | | | | | New instruction format 'ivABC' (a variant of iABC where parameter vC has 10 bits) allows constructors of up to 1024 elements to be coded without EXTRAARG.
* 'isIT'/'isOT' turned from macros to functionsRoberto Ierusalimschy2024-06-271-1/+3
|
* Flexible limit for use of registers by constructorsRoberto Ierusalimschy2024-06-271-1/+1
| | | | | | | | Instead of a fixed limit of 50 registers (which, in a bad worst case, can limit the nesting of constructors to 5 levels), the compiler computes an individual limit for each constructor based on how many registers are available when it runs. This limit then controls the frequency of SETLIST instructions.
* Clearer code for controlling maximum registersRoberto Ierusalimschy2024-06-261-5/+1
| | | | Plus, added a test to check that limit.
* Cleaning of llimits.hRoberto Ierusalimschy2024-06-201-3/+3
| | | | | | Several definitions that don't need to be "global" (that is, that concerns only specific parts of the code) moved out of llimits.h, to more appropriate places.
* Bug: overlapping assignmentsRoberto Ierusalimschy2024-06-121-2/+4
| | | | | ISO C forbids assignment of a union field to another field of the same union.
* 'luaH_get' functions return tag of the resultRoberto Ierusalimschy2024-03-211-2/+3
| | | | | | | Undoing previous commit. Returning TValue increases code size without any visible gains. Returning the tag is a little simpler than returning a special code (HOK/HNOTFOUND) and the tag is useful by itself in some cases.
* 'luaH_get' functions return 'TValue'Roberto Ierusalimschy2024-03-181-3/+2
| | | | | | Instead of receiving a parameter telling them where to put the result of the query, these functions return the TValue directly. (That is, they return a structure.)
* Merge branch 'newarray' into nextversionRoberto Ierusalimschy2023-11-071-4/+4
|\
| * Merge branch 'master' into newarrayRoberto Ierusalimschy2023-11-031-14/+15
| |\
| * | Some cleaning in the new table APIRoberto Ierusalimschy2023-05-161-4/+4
| | |
* | | Field 'Proto.is_vararg' uses only one bitRoberto Ierusalimschy2023-08-301-2/+2
| |/ |/| | | | | So that the other bits can be used for other purposes.
* | Several functions turned 'static'Roberto Ierusalimschy2023-05-221-11/+11
| | | | | | | | | | Several functions that were already being used only inside their own file have been declared as 'static'.
* | DetailsRoberto Ierusalimschy2023-05-151-3/+4
|/ | | | | - Better comments about short strings in opcodes. - luaH_newkey made static.
* More orderliness in casts of enumerationsRoberto Ierusalimschy2023-03-271-14/+41
|
* More regularity in uses of enums in 'lcode.c'Roberto Ierusalimschy2023-03-241-18/+18
|
* Factoring out common parts of 'codearith' and 'codebitwise'Roberto Ierusalimschy2022-05-061-25/+33
|
* Bug: Wrong code generation in bitwise operationsRoberto Ierusalimschy2022-04-251-6/+10
|
* DetailRoberto Ierusalimschy2022-01-031-1/+1
| | | | Warnings with clang when using long double for Lua floats.
* Changes in cache for function constantsRoberto Ierusalimschy2021-03-301-8/+26
| | | | | | | | In 'lcode.c', when adding constants to the list of constants of a function, integers represent themselves in the cache and floats with integral values get a small delta to avoid collision with integers. (This change avoids creating artificial addresses; the old implementation converted integers to pointers to index the cache.)
* Don't use tointegerns when luaV_tointegerns will doRoberto Ierusalimschy2021-02-241-1/+2
| | | | | | | | Some places don't need the "fast path" macro tointegerns, either because speed is not essential (lcode.c) or because the value is not supposed to be an integer already (luaV_equalobj and luaG_tointerror). Moreover, luaV_equalobj should always use F2Ieq, even if Lua is compiled to "round to floor".
* Optimization/simplification of 'getbaseline'Roberto Ierusalimschy2021-02-021-2/+2
| | | | | By producing absolute line information at regular intervals, a simple division can compute the correct entry for a given instruction.
* Optimizations for line hookRoberto Ierusalimschy2021-01-281-9/+0
| | | | | The function 'changedline' tries harder to avoid calling 'luaG_getfuncline' plus small changes in the use of 'L->oldpc'.
* Changes in the API of 'luaH_set' and related functionsRoberto Ierusalimschy2020-12-041-2/+6
| | | | | Functions to set values in a table (luaH_set, luaH_newkey, etc.) receive the new value, instead of returning a slot where to put the value.
* DetailsRoberto Ierusalimschy2020-12-021-3/+3
| | | | Names in the parser and other details that do not change actual code.
* DetailsRoberto Ierusalimschy2020-10-301-5/+9
| | | | | | - small corrections in the manual - ldo.c: 'docall' -> 'ccall' ('docall' already used in 'lua.c') - comments
* Fixed "conceptual" bug in 'luaK_setreturns'Roberto Ierusalimschy2020-03-021-4/+3
| | | | | | | | This function was computing invalid instruction addresses when the expression was not a multi-return instruction. (Virtually all machines don't raise errors when computing an invalid address, as long as the address is not accessed, but this computation is undefined behavior in ISO C.)
* DetailsRoberto Ierusalimschy2020-02-271-10/+5
| | | | | Several details in code (e.g., moving a variable to the most inner scope that encloses its uses), comments, parameter names, extra tests.
* OP_LOADFALSE broken in two instructionsRoberto Ierusalimschy2020-02-111-4/+4
|
* Clearer distinction between types and tagsRoberto Ierusalimschy2020-01-311-6/+6
| | | | | LUA_T* represents only types; tags (types + Variants) are represented by LUA_V* constants.
* Changed internal representation of booleansRoberto Ierusalimschy2020-01-061-15/+35
| | | | | | | Instead of an explicit value (field 'b'), true and false use different tag variants. This avoids reading an extra field and results in more direct code. (Most code that uses booleans needs to distinguish between true and false anyway.)
* Comments (mosty typos)Roberto Ierusalimschy2019-12-301-1/+1
|
* Using an enumeration for float->integer coercion modesRoberto Ierusalimschy2019-12-051-2/+2
|
* Fixed warnings from Keil compilerRoberto Ierusalimschy2019-10-251-1/+1
|
* Fixed a warning and other minor issuesRoberto Ierusalimschy2019-10-041-2/+2
| | | | Fixed some minor issues from the feedback for 5.4-beta rc1.
* Janitorial workRoberto Ierusalimschy2019-10-011-32/+29
| | | | | | | - Several details in 'lcode.c' - A few more tests for code generation - Bug in assert in 'lcode.c' ("=" x "==") - Comments in 'lopcodes.h' and 'ltable.c'
* Subtraction of small constant integers optimized with OP_ADDIRoberto Ierusalimschy2019-09-241-27/+41
|
* Janitorial work in 'lcode.c'Roberto Ierusalimschy2019-09-191-33/+25
|
* Simplification in the call to 'constfolding'Roberto Ierusalimschy2019-09-111-15/+11
|
* Removed arithmetic opcodes with immediate operandRoberto Ierusalimschy2019-09-101-8/+8
| | | | | | | The difference in performance between immediate operands and K operands does not seem to justify all those extra opcodes. We only keep OP_ADDI, due to its ubiquity and because the difference is a little more relevant. (Later, OP_SUBI will be implemented by OP_ADDI, negating the constant.)
* Use of 'MMBIN' opcodes extended to shift operatorsRoberto Ierusalimschy2019-08-281-11/+15
| | | | Plus, this commit removes useless 'tm' parameters in 'op_*' macros.
* First version of OP_MMBIN opcodesRoberto Ierusalimschy2019-08-271-19/+27
| | | | | | | | | | | In arithmetic/bitwise operators, the call to metamethods is made in a separate opcode following the main one. (The main opcode skips this next one when the operation succeeds.) This change reduces slightly the size of the binary and the complexity of the arithmetic/bitwise opcodes. It also simplfies the treatment of errors and yeld/resume in these operations, as there are much fewer cases to consider. (Only OP_MMBIN/OP_MMBINI/OP_MMBINK, instead of all variants of all arithmetic/bitwise operators.)
* Fixed small issue with constant propagationRoberto Ierusalimschy2019-07-171-3/+15
| | | | | | | | Constants directly assigned to other constants were not propagating: For instance, in local <const> k1 = 10 local <const> k2 = k1 'k2' were not treated as a compile-time constant.
* New kind of expression VKSTRRoberto Ierusalimschy2019-07-171-10/+25
| | | | | | | String literal expressions have their own kind VKSTR, instead of the generic VK. This allows strings to "cross" functions without entering their constant tables (e.g., if they are used only by some nested function).
* Micro optimization in OP_RETURN and OP_TAILCALLRoberto Ierusalimschy2019-07-161-4/+4
| | | | | Many functions are vararg but create no upvalues, so it is better to separate the tests for these two kinds of "extra work".
* Unification of size representation in OP_NEWTABLE and OP_SETLISTRoberto Ierusalimschy2019-07-151-12/+28
| | | | | | Opcodes OP_NEWTABLE and OP_SETLIST use the same representation to store the size of the array part of a table. This new representation can go up to 2^33 (8 + 25 bits).
* OP_NEWTABLE keeps exact size of arraysRoberto Ierusalimschy2019-07-121-3/+3
| | | | | | OP_NEWTABLE is followed by an OP_EXTRAARG, so that it can keep the exact size of the array part of the table to be created. (Functions 'luaO_int2fb'/'luaO_fb2int' were removed.)
* First implementation of constant propagationRoberto Ierusalimschy2019-07-121-1/+55
| | | | | Local constant variables initialized with compile-time constants are optimized away from the code.
* Towards constant propagationRoberto Ierusalimschy2019-07-101-2/+2
| | | | | | | This commit detaches the number of active variables from the number of variables in the stack, during compilation. Soon, compile-time constants will be propagated and therefore will not exist during run time (in the stack).
* New implementation for constantsRoberto Ierusalimschy2019-07-091-47/+13
| | | | | | | | | VLOCAL expressions keep a reference to their corresponding 'Vardesc', and 'Upvaldesc' (for upvalues) has a field 'ro' (read-only). So, it is easier to check whether a variable is read-only. The decoupling in VLOCAL between 'vidx' ('Vardesc' index) and 'sidx' (stack index) should also help the forthcoming implementation of compile-time constant propagation.
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-12-26 14:01:35 +0000