nsieve benchmark and MSDN info about mem allocation strategies

[cobold]

Greetings,

just for fun I have coded this nsieve-prog. Initially I used GlobalAlloc/GlobalFree to allocate memory for the byte-array.
Our friends at MSDN say:

GlobalAlloc Function
Allocates the specified number of bytes from the heap. .....
New applications should use the heap functions unless documentation states that a global function should be used. .....

Therefore, GlobalAlloc and LocalAlloc have greater overhead than HeapAlloc.

colbold's nsieve.asm with GlobalAlloc:
C:\Asm>nsieve 10
Primes up to 10240000   679461
Primes up to  5120000   356244
Primes up to  2560000   187134
3094 ms
colbold's nsieve.asm with HeapAlloc:
Primes up to 10240000   679461
Primes up to  5120000   356244
Primes up to  2560000   187134
3124 ms

So, using GlobalAlloc is 30 ms faster than HeapAlloc (10 ms per Alloc/Free pair), whereas MSDN states otherwise?

What do you experts out there advise as memory allocation strategy?

P.S.:

Code Select Expand


; The Computer Language Shootout
; http://shootout.alioth.debian.org/

;Contributed by Klaus Kolba
;Nsieve

include \masm32\include\masm32rt.inc

.DATA
    cmd     db 128 dup(?)
    fmtStr  db "Primes up to %8u %8u",13,10,0
align 4
    count   dd ?
    hHeap   dd ?
   
.CODE
nsieve PROC
    push ebx
    push ecx

    invoke GlobalAlloc,GMEM_FIXED or GMEM_ZEROINIT,edi
    ;invoke GetProcessHeap
    ;mov hHeap,eax
    ;invoke HeapAlloc,eax,HEAP_ZERO_MEMORY,edi
    mov esi,eax

    xor edx,edx;mov count,0
    mov ebx,2
    ALIGN 4
    .WHILE ebx < edi
        .IF BYTE PTR [esi+ebx] == 0
            inc edx;count
            mov eax,ebx
            shl eax,1
            mov ecx,eax
            .WHILE ecx < edi
                mov BYTE PTR [esi+ecx],1
                add ecx,ebx
            .ENDW
        .ENDIF
        inc ebx
    .ENDW
    invoke crt_printf,ADDR fmtStr,edi,edx   ;edx=count
    invoke GlobalFree,esi
    ;invoke HeapFree,hHeap,HEAP_NO_SERIALIZE,esi
    pop ecx
    pop ebx
    ret
nsieve ENDP
   
start:
    ;invoke GetCurrentProcess
    ;invoke SetPriorityClass,eax,HIGH_PRIORITY_CLASS
   
    invoke GetTickCount
    push eax
   
    invoke GetCL,1,ADDR cmd
    invoke atodw,ADDR cmd
    mov ecx,eax

    xor ebx,ebx
    ALIGN 4
    .WHILE ebx <= 2
        mov eax,10000
        sub ecx,ebx
        shl eax,cl
        add ecx,ebx
        mov edi,eax
        invoke nsieve
        inc ebx
    .ENDW

    invoke GetTickCount
    pop ebx
    sub eax,ebx
    print str$(eax)," ms"
    invoke ExitProcess,0
end start