Re: [Benchmark] Chain calls

["Alexander Gladysh" <agladysh@...>]

On Tue, Nov 25, 2008 at 7:19 AM, KHMan <keinhong@gmail.com> wrote:
> Alexander Gladysh wrote:
>>
>> On Mon, Nov 17, 2008 at 3:38 PM, Peter Cawley <lua@corsix.org> wrote:
>>>
>>> It may be worth looking at the generated Lua opcodes for these benchmarks
>>> in
>>> order to easier see the differences in what is happening in each. For
>>> example, return true v.s return nil are loadbool,return vs.
>>> loadnil,return.
>>> Then looking at the VM code for these operations, either in C or as the
>>> assembled output of the C, might make it clearer. Of course, this won't
>>> help
>>> with explaining the luajit results, as it skips the VM when JITing.
>>
>> Sorry for the late reply.
>>
>> Opcode listing (via luac -l -l) is indeed very helpful. Chaining calls
>> use less resources, since they do not require extra MOVE opcodes:
>>
>> local function chain_local()
>>  local chain = chain
>>  chain () () () () () () () () () () -- 10 calls
>> end
>
> [snip]
>
> Trouble is, does this pattern ever appear in normal code? I find it
> difficult to believe that calling a function that returns a function to be
> called, and so on, chained in this manner, could happen in real-world code.

Well, it does in mine. I often use this idiom for heavy concatenation
code (for one thing, you may easily replace buffering cat() with
io.write()):

local buf = {}

local function cat(s)
  buf[#buf + 1] = s
  return cat
end

local function test()
  cat "one" "two" "four" "five"
end

function <chainstr.lua:8,10> (10 instructions, 40 bytes at 0x100eb0)
0 params, 2 slots, 1 upvalue, 0 locals, 4 constants, 0 functions
	1	[9]	GETUPVAL 	0 0	; cat
	2	[9]	LOADK    	1 -1	; "one"
	3	[9]	CALL     	0 2 2
	4	[9]	LOADK    	1 -2	; "two"
	5	[9]	CALL     	0 2 2
	6	[9]	LOADK    	1 -3	; "four"
	7	[9]	CALL     	0 2 2
	8	[9]	LOADK    	1 -4	; "five"
	9	[9]	CALL     	0 2 1
	10	[10]	RETURN   	0 1

Alexander.