A few errors to deal with
in [ASM]
|
Prev: Lolling at programmers, how many ways are there to create a bitmask? ;) :)
Next: Lolling at programmers, how many ways are there to create abitmask ? ;) :)
From: Branimir Maksimovic on 31 May 2010 08:35 On Mon, 31 May 2010 13:34:10 +0200 Branimir Maksimovic <bmaxa(a)hotmail.com> wrote: > > On Sun, 30 May 2010 05:16:03 -0400 > > Frank Kotler <fbkotler(a)myfairpoint.net> wrote: > > > > > > > > My output from this is "21C" (with a bunch of zeros in front). > > > With the "five byte padding" uncommented, it goes to "220". All > > > we're "timing" is push edx/push eax/cpuid... is cpuid sensitive to > > > alignment??? I would expect that if five bytes changes it, one > > > byte would, too - but it doesn't (your mileage may vary)... > > > > bmaxa(a)maxa:~/fasm/test$ fasm ttest.asm > flat assembler version 1.68 (16384 kilobytes memory) > 2 passes, 230 bytes. > bmaxa(a)maxa:~/fasm/test$ ./ttest > 000000000000019E > 00000000000001A7 > 000000000000019E > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 000000000000019E > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > 00000000000001A7 > bmaxa(a)maxa:~/fasm/test$ cat ttest.asm > ; fasm myprog.asm > ; > ; from Branimir Maksimovic > ; bugfixes from Nathan Baker > ; cruft from fbk :) > > format ELF executable > > segment writeable executable > > entry $ > > mov ebx, xtbl > > ;nop > ;nop > ;nop > ;nop > ;nop > ;nop > > mov ecx,16 > l1: > push ecx > > cpuid > rdtsc > push edx > push eax > > das > push eax > pop eax > push eax > pop eax > > cpuid > rdtsc > > pop ebx > sub eax, ebx > pop ecx > sub edx, ecx > > mov edi, ascbuf > call u64toha > > mov ecx, ascbuf > mov edx, 17 > mov ebx, 1 > mov eax, 4 > int 80h > > pop ecx > loop l1 > > exit: > mov eax, 1 > mov ebx,0 > int 80h > > xtbl db 30h,31h,32h,33h,34h,35h,36h,37h,38h,39h,41h,42h, \ > 43h,44h,45h,46h > > u64toha: > add edi, 15 > mov ebx,xtbl > mov cl, 16 > std > l2: > mov ch,al > and al,0xf > xlatb > stosb > mov al,ch > ; shrd edx,eax,4 > shrd eax,edx,4 > shr edx, 4 > dec cl > jz e1 > ; mov byte[edi], ',' > ; inc edi > jmp l2 > > e1: > cld > ret > > ascbuf db 17 dup (0xa) > bmaxa(a)maxa:~/fasm/test$ > > Greets! > bmaxa(a)maxa:~/fasm/test$ cat ttest.asm ; fasm myprog.asm ; ; from Branimir Maksimovic ; bugfixes from Nathan Baker ; cruft from fbk :) format ELF executable segment writeable executable entry $ mov ebx, xtbl ;nop ;nop ;nop ;nop ;nop ;nop mov ecx,16 l1: push ecx cpuid rdtsc push edx push eax das push eax pop eax push eax pop eax cpuid rdtsc pop ebx sub eax, ebx pop ecx sub edx, ecx mov edi, ascbuf call u64toha mov ecx, ascbuf mov edx, 17 mov ebx, 1 mov eax, 4 int 80h pop ecx dec ecx jnz l1 exit: mov eax, 1 mov ebx,0 int 80h xtbl db 30h,31h,32h,33h,34h,35h,36h,37h,38h,39h,41h,42h, \ 43h,44h,45h,46h u64toha: add edi, 15 mov ebx,xtbl mov cl, 16 std l2: mov ch,al and al,0xf xlatb stosb mov al,ch ; shrd edx,eax,4 shrd eax,edx,4 shr edx, 4 dec cl jz e1 ; mov byte[edi], ',' ; inc edi jmp l2 e1: cld ret ascbuf db 17 dup (0xa) bmaxa(a)maxa:~/fasm/test$ Greets! -- http://maxa.homedns.org/ Sometimes online sometimes not Svima je "dozvoljeno" biti idiot i > mrak, ali samo neki to odaberu,
From: Branimir Maksimovic on 31 May 2010 08:39 On Mon, 31 May 2010 14:35:01 +0200 Branimir Maksimovic <bmaxa(a)hotmail.com> wrote: > > format ELF executable > > segment writeable executable > > entry $ > > mov ebx, xtbl > > ;nop > ;nop > ;nop > ;nop > ;nop > ;nop > > mov ecx,16 > l1: > push ecx > > cpuid > rdtsc > push edx > push eax > > das > push eax > pop eax > push eax > pop eax > > cpuid > rdtsc > > pop ebx > sub eax, ebx > pop ecx > sub edx, ecx > > mov edi, ascbuf > call u64toha > > mov ecx, ascbuf > mov edx, 17 > mov ebx, 1 > mov eax, 4 > int 80h > > pop ecx > dec ecx > jnz l1 > > exit: > mov eax, 1 > mov ebx,0 > int 80h > > xtbl db 30h,31h,32h,33h,34h,35h,36h,37h,38h,39h,41h,42h, \ > 43h,44h,45h,46h > > u64toha: > add edi, 15 > mov ebx,xtbl > mov cl, 16 > std > l2: > mov ch,al > and al,0xf > xlatb > stosb > mov al,ch > ; shrd edx,eax,4 > shrd eax,edx,4 > shr edx, 4 > dec cl > jz e1 > ; mov byte[edi], ',' > ; inc edi > jmp l2 > > e1: > cld > ret > > ascbuf db 17 dup (0xa) > bmaxa(a)maxa:~/fasm/test$ > > Greets! bmaxa(a)maxa:~/fasm/test$ fasm ttest.asm flat assembler version 1.68 (16384 kilobytes memory) 2 passes, 231 bytes. bmaxa(a)maxa:~/fasm/test$ ./ttest 0000000000000183 0000000000000183 0000000000000183 000000000000017A 000000000000017A 000000000000018C 000000000000017A 000000000000017A 000000000000017A 0000000000000183 0000000000000183 000000000000017A 000000000000017A 0000000000000183 0000000000000183 0000000000000183 bmaxa(a)maxa:~/fasm/test$ fasm ttest.asm flat assembler version 1.68 (16384 kilobytes memory) 2 passes, 237 bytes. bmaxa(a)maxa:~/fasm/test$ ./ttest 00000000000001B9 0000000000000183 000000000000017A 000000000000017A 0000000000000183 0000000000000183 000000000000017A 0000000000000183 000000000000017A 0000000000000183 000000000000017A 000000000000017A 000000000000017A 000000000000017A 0000000000000183 000000000000017A bmaxa(a)maxa:~/fasm/test$ cat ttest.asm ; fasm myprog.asm ; ; from Branimir Maksimovic ; bugfixes from Nathan Baker ; cruft from fbk :) format ELF executable segment writeable executable entry $ mov ebx, xtbl nop nop nop nop nop nop mov ecx,16 l1: push ecx cpuid rdtsc push edx push eax das push eax pop eax push eax pop eax cpuid rdtsc pop ebx sub eax, ebx pop ecx sub edx, ecx mov edi, ascbuf call u64toha mov ecx, ascbuf mov edx, 17 mov ebx, 1 mov eax, 4 int 80h pop ecx dec ecx jnz l1 exit: mov eax, 1 mov ebx,0 int 80h xtbl db 30h,31h,32h,33h,34h,35h,36h,37h,38h,39h,41h,42h, \ 43h,44h,45h,46h u64toha: add edi, 15 mov ebx,xtbl mov cl, 16 std l2: mov ch,al and al,0xf xlatb stosb mov al,ch ; shrd edx,eax,4 shrd eax,edx,4 shr edx, 4 dec cl jz e1 ; mov byte[edi], ',' ; inc edi jmp l2 e1: cld ret ascbuf db 17 dup (0xa) bmaxa(a)maxa:~/fasm/test$ SO CATCH WAS IN LOOP INSTRUCTION ;) Greets! -- http://maxa.homedns.org/ Sometimes online sometimes not Svima je "dozvoljeno" biti idiot i > mrak, ali samo neki to odaberu,
From: wolfgang kern on 1 Jun 2010 12:51 Frank Kotler wrote: .... > ; five bytes here changes the timing > ;mov ebx, xtbl > > ;nop > ;nop > ;nop > ;nop > ;nop > ;nop ; six bytes changes it back > > mov ecx,16 > l1: > push ecx > > ; serialize CPU and get start time > cpuid > rdtsc > push edx > push eax > > ; code to be timed > ;-------------- > ;das > ;push eax > ;pop eax > ;push eax > ;pop eax > ;-------------- **** > ; serialize cpu and get end time > cpuid this 'second serializing' may be the main amount of cycles you get. > rdtsc > > ; calculate difference > pop ebx > sub eax, ebx > pop ecx > sub edx, ecx .... > My output from this is "21C" (with a bunch of zeros in front). With the > "five byte padding" uncommented, it goes to "220". > All we're "timing" is push edx/push eax/cpuid... is cpuid sensitive to > alignment??? No, but it serializes (timing depends on previous pipe/cache/prefetch...) > I would expect that if five bytes changes it, one byte would, too > - but it doesn't (your mileage may vary)... these added NOPs may just alter the code prefetch or cache status? > That first output you posted - varying between "1" and "A" - was that for > an empty loop, or was that with "das" in there? I'm getting consistent > results (if sometimes puzzling) for all 16 iterations... with anything but > "das"... An empty test-loop would contain 2 times MOV r,r (or PUSH r) and one RDTSC. RDTSC timing is machine specific (5...else cycles). I think to have posted this several times already: MOV esi,result_buf ;16 bytes needed yet here MOV byte[loop],01 CLI L1: ;PUSH esi ;if desired/required CPUID RDTSC MOV ebx,eax ;or: PUSH eax MOV ecx,edx ;or: PUSH edx ... ;code under test (must preserve what's needed) RDTSC ;without serialising yet !!! SUB eax,ebx ;or: SUB eax,[esp+4] SBB edx,ecx ;or: SBB edx,[esp] ;or: ADD esp,8 ;POP esi MOV [esi],eax ; MOV [esi+4],edx ;store 64-bit cycle-count ADD esi,8 DEC byte [loop] JNS L1 ;loop the above just one more time STI A first run (I intentionally avoid the term 'iterate' here) would show a cycle count which implies cache-burst-reads. The second test-run follows immediate still with IRQs disabled. One million iterations may just measure OS-noise from IRQ- taskswitching or whatsoever an OS may do behind your back. I measure 6..7 cycles (almost just the RDTSC itself) on an empty (single NOP) test one my current machine with the MOV- version while the PUSH-variant takes 1..2 cycles more. Usually I ignore the first 64-bit result (caching time) for code variants compare, but the info there is very useful for proper code alignment. Note: this method shouldn't be used on huge code parts, too long disabled IRQs can result in stuck hardware. My timing check on DAS seem to confirm the AMD-books, it takes 8 cycles regardless if eax were altered before or not, but timing become worse if several of these 'vectored' instructions are within one codefetch range. __ wolfgang
From: Mint on 10 Jun 2010 15:12
On May 28, 5:42 am, Frank Kotler <fbkot...(a)myfairpoint.net> wrote: > Branimir Maksimovic wrote: > > On Thu, 27 May 2010 22:50:22 -0400 > > Frank Kotler <fbkot...(a)myfairpoint.net> wrote: > Andy, are you still getting 340 cycles for any block of code? > Best, > Frank Sorry for the late response. This is my last code. The results are "relatively" close to what they should be. Even running from a boot disk, I get 55 cycles. Here's the .exe. Andy begin 0666 timeit.exe M35KT`0T``0`@`$$`__^``0`$```1`0``'@````$`$@$````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M```````````````````````````````````````````````````````````` M``````````````````````!5B^Q3BUX$@#\`=`-#Z_B+PRM&!%O)PU6+[%!3 M45+_=(a)3HW/^#Q`*+R+L!`(M6!+1`S2%:65M8R<.T`<T6=`:T`,T6Z_2T`<T6 M=/JT`,T64+0!S19T!K0`S1;K]%C#58OL4%-14E=FBT8$BWX(9H7`>0=F]]C& M!2U'9KL*````,\EF,])F]_-2068+P'7S6`0PB`5'XOC&!0!?6EE;6,G#!E.X M0`".P+ML`&8FBP=;!\-5B^SHZ?]FB]!F`U8$Z-__9CO"<OC)PY"0D)"0D)"0 MD)"0D)"0D)!5B^Q34HM6!#/`B]C1ZA/#(])U^%I;R<.0D)"0D)"0D)"0D)"0 MD)"04S/`4HO8$\/1ZG7Z6A/#6\.X:P&.V(S3*]C!XP2.T`/C9F@R````Z(?_ M@\0$^F;'!B0`_____V;'!B@`_____V;'!BP`_____V;'!C``_____V;'!C0` M9````&;'!C@`9````&939C/`#Z)F,\`/HF8SP`^B9EN0D)"0D)"0D)"0D)"0 MD)"0D)"0D)"0D)"09E-F,\`/H@\Q9EMF4F909E-F,\`/HF9;9E-F,\`/H@\Q M9EMF668KP6999AO19CL6,`!U!68[!BP`<PEFHRP`9HD6,`!F(a)RXX``%UM&;' M!C@`9````)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D&939C/`#Z(/,69; M9E)F4&939C/`#Z)F6V939C/`#Z(/,69;9EEF*\%F668;T68[%B@`=05F.P8D M`',)9J,D`&:)%B@`9H,N.``!=;1FH20`9BL&+`!FBQ8H`&8;%C``^V@&`&90 MZ/;]@\0&:`8`Z*;]@\0":#P`Z)W]@\0"^F;'!B0`_____V;'!B@`_____V;' M!BP`_____V;'!C``_____V;'!C0`9````&;'!C@`9````&939C/`#Z)F,\`/ MHF8SP`^B9EN0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"09E-F,\`/H@\Q9EMF M4F909E-F,\`/HF9;9E-F,\`/H@\Q9EMF668KP6999AO19CL6,`!U!68[!BP` M<PEFHRP`9HD6,`!F(a)RXX``%UM&;'!C@`9````)"0D)"0D)"0D)"0D)"0D)"0 MD)"0D)"0D)"0D&939C/`#Z(/,69;9E)F4&939C/`#Z)F6V:XT/L!`&:YH/<# M`&:99O?Y9E-F,\`/H@\Q9EMF668KP6999AO19CL6*`!U!68[!B0`<PEFHR0` M9HD6*`!F(a)RXX``%UHV:A)`!F*P8L`&:+%B@`9AL6,`#[:`8`9E#HE?R#Q`9H M!@#H1?R#Q`)H30#H//R#Q`+Z9L<&)`#_____9L<&*`#_____9L<&+`#_____ M9L<&,`#_____9L<&-`!D````9L<&.`!D````9E-F,\`/HF8SP`^B9C/`#Z)F M6Y"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D&939C/`#Z(/,69;9E)F4&939C/` M#Z)F6V939C/`#Z(/,69;9EEF*\%F668;T68[%C``=05F.P8L`',)9J,L`&:) M%C``9H,N.``!=;1FQP8X`&0```"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0 MD)!F4V8SP`^B#S%F6V929E!F4V8SP`^B9EMF4V8SP`^B#S%F6V999BO!9EEF M&]%F.Q8H`'4%9CL&)`!S"6:C)`!FB18H`&:#+C@``76T9J$D`&8K!BP`9HL6 M*`!F&Q8P`/MH!@!F4.A&^X/$!F@&`.CV^H/$`FA=`.CM^H/$`OIFQP8D`/__ M__]FQP8H`/____]FQP8L`/____]FQP8P`/____]FQP8T`&0```!FQP8X`&0` M``!F4V8SP`^B9C/`#Z)F,\`/HF9;D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0 MD&939C/`#Z(/,69;9E)F4&939C/`#Z)F6V939C/`#Z(/,69;9EEF*\%F668; MT68[%C``=05F.P8L`',)9J,L`&:)%C``9H,N.``!=;1FQP8X`&0```"0D)"0 MD)"0D)"0D)"0D)"0D)"0D)"0D)"0D)!F4V8SP`^B#S%F6V929E!F4V8SP`^B M9EMFN-#[`0!FN:#W`P!FF6;W^6939C/`#Z(/,69;9EEF*\%F668;T68[%B@` M=05F.P8D`',)9J,D`&:)%B@`9H,N.``!=:-FH20`9BL&+`!FBQ8H`&8;%C`` M^V@&`&90Z.7Y@\0&:`8`Z)7Y@\0":&X`Z(SY@\0"^F;'!B0`_____V;'!B@` M_____V;'!BP`_____V;'!C``_____V;'!C0`9````&;'!C@`9````&939C/` M#Z)F,\`/HF8SP`^B9EN0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)!F4V8SP`^B M#S%F6V929E!F4V8SP`^B9EMF4V8SP`^B#S%F6V999BO!9EEF&]%F.Q8P`'4% M9CL&+`!S"6:C+`!FB18P`&:#+C@``76T9L<&.`!D````D)"0D)"0D)"0D)"0 MD)"0D)"0D)"0D)"0D)"09E-F,\`/H@\Q9EMF4F909E-F,\`/HF9;9E-F,\`/ MH@\Q9EMF668KP6999AO19CL6*`!U!68[!B0`<PEFHR0`9HD6*`!F(a)RXX``%U MM&:A)`!F*P8L`&:+%B@`9AL6,`#[:`8`9E#HEOB#Q`9H!@#H1OB#Q`)H?@#H M/?B#Q`+Z9L<&)`#_____9L<&*`#_____9L<&+`#_____9L<&,`#_____9L<& M-`!D````9L<&.`!D````9E-F,\`/HF8SP`^B9C/`#Z)F6Y"0D)"0D)"0D)"0 MD)"0D)"0D)"0D)"0D)!F4V8SP`^B#S%F6V929E!F4V8SP`^B9EMF4V8SP`^B M#S%F6V999BO!9EEF&]%F.Q8P`'4%9CL&+`!S"6:C+`!FB18P`&:#+C@``76T M9L<&.`!D````D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"09E-F,\`/H@\Q M9EMF4F909E-F,\`/HF9;9KC0^P$`9KF@]P,`9IEF]_EF4V8SP`^B#S%F6V99 M9BO!9EEF&]%F.Q8H`'4%9CL&)`!S"6:C)`!FB18H`&:#+C@``76C9J$D`&8K M!BP`9HL6*`!F&Q8P`/MH!@!F4.@U]X/$!F@&`.CE]H/$`FB/`.C<]H/$`OIF MQP8D`/____]FQP8H`/____]FQP8L`/____]FQP8P`/____]FQP8T`&0```!F MQP8X`&0```!F4V8SP`^B9C/`#Z)F,\`/HF9;D)"0D)"0D)"0D)"0D)"0D)"0 MD)"0D)"09E-F,\`/H@\Q9EMF4F909E-F,\`/HF9;9E-F,\`/H@\Q9EMF668K MP6999AO19CL6,`!U!68[!BP`<PEFHRP`9HD6,`!F(a)RXX``%UM&;'!C@`9``` M`)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D&939C/`#Z(/,69;9E)F4&93 M9C/`#Z)F6V939C/`#Z(/,69;9EEF*\%F668;T68[%B@`=05F.P8D`',)9J,D M`&:)%B@`9H,N.``!=;1FH20`9BL&+`!FBQ8H`&8;%C``^V@&`&90Z.;U@\0& M:`8`Z);U@\0":)\`Z(WU@\0"^F;'!B0`_____V;'!B@`_____V;'!BP`____ M_V;'!C``_____V;'!C0`9````&;'!C@`9````&939C/`#Z)F,\`/HF8SP`^B M9EN0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"09E-F,\`/H@\Q9EMF4F909E-F M,\`/HF9;9E-F,\`/H@\Q9EMF668KP6999AO19CL6,`!U!68[!BP`<PEFHRP` M9HD6,`!F(a)RXX``%UM&;'!C@`9````)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0 MD)"0D&939C/`#Z(/,69;9E)F4&939C/`#Z)F6V:XT/L!`&:YH/<#`&:99O?Y M9E-F,\`/H@\Q9EMF668KP6999AO19CL6*`!U!68[!B0`<PEFHR0`9HD6*`!F M(a)RXX``%UHV:A)`!F*P8L`&:+%B@`9AL6,`#[:`8`9E#HA?2#Q`9H!@#H-?2# MQ`)HL`#H+/2#Q`+Z9L<&)`#_____9L<&*`#_____9L<&+`#_____9L<&,`#_ M____9L<&-`!D````9L<&.`!D````9E-F,\`/HF8SP`^B9C/`#Z)F6Y"0D)"0 MD)"0D)"0D)"0D)"0D)"0D)"0D&939C/`#Z(/,69;9E)F4&939C/`#Z)F6V93 M9C/`#Z(/,69;9EEF*\%F668;T68[%C``=05F.P8L`',)9J,L`&:)%C``9H,N M.``!=;1FQP8X`&0```"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)!F4V8S MP`^B#S%F6V929E!F4V8SP`^B9EMF4V8SP`^B#S%F6V999BO!9EEF&]%F.Q8H M`'4%9CL&)`!S"6:C)`!FB18H`&:#+C@``76T9J$D`&8K!BP`9HL6*`!F&Q8P M`/MH!@!F4.@V\X/$!F@&`.CF\H/$`FC``.C=\H/$`OIFQP8D`/____]FQP8H M`/____]FQP8L`/____]FQP8P`/____]FQP8T`&0```!FQP8X`&0```!F4V8S MP`^B9C/`#Z)F,\`/HF9;D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D&939C/` M#Z(/,69;9E)F4&939C/`#Z)F6V939C/`#Z(/,69;9EEF*\%F668;T68[%C`` M=05F.P8L`',)9J,L`&:)%C``9H,N.``!=;1FQP8X`&0```"0D)"0D)"0D)"0 MD)"0D)"0D)"0D)"0D)"0D)!F4V8SP`^B#S%F6V929E!F4V8SP`^B9EMFN-#[ M`0!FN:#W`P!FF6;W^6939C/`#Z(/,69;9EEF*\%F668;T68[%B@`=05F.P8D M`',)9J,D`&:)%B@`9H,N.``!=:-FH20`9BL&+`!FBQ8H`&8;%C``^V@&`&90 MZ-7Q@\0&:`8`Z(7Q@\0":-$`Z'SQ@\0"^F;'!B0`_____V;'!B@`_____V;' M!BP`_____V;'!C``_____V;'!C0`9````&;'!C@`9````&939C/`#Z)F,\`/ MHF8SP`^B9EN0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)!F4V8SP`^B#S%F6V92 M9E!F4V8SP`^B9EMF4V8SP`^B#S%F6V999BO!9EEF&]%F.Q8P`'4%9CL&+`!S M"6:C+`!FB18P`&:#+C@``76T9L<&.`!D````D)"0D)"0D)"0D)"0D)"0D)"0 MD)"0D)"0D)"09E-F,\`/H@\Q9EMF4F909E-F,\`/HF9;9E-F,\`/H@\Q9EMF M668KP6999AO19CL6*`!U!68[!B0`<PEFHR0`9HD6*`!F(a)RXX``%UM&:A)`!F M*P8L`&:+%B@`9AL6,`#[:`8`9E#HAO"#Q`9H!@#H-O"#Q`)HX0#H+?"#Q`+Z M9L<&)`#_____9L<&*`#_____9L<&+`#_____9L<&,`#_____9L<&-`!D```` M9L<&.`!D````9E-F,\`/HF8SP`^B9C/`#Z)F6Y"0D)"0D)"0D)"0D)"0D)"0 MD)"0D)"0D)!F4V8SP`^B#S%F6V929E!F4V8SP`^B9EMF4V8SP`^B#S%F6V99 M9BO!9EEF&]%F.Q8P`'4%9CL&+`!S"6:C+`!FB18P`&:#+C@``76T9L<&.`!D M````D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"09E-F,\`/H@\Q9EMF4F90 M9E-F,\`/HF9;9KC0^P$`9KF@]P,`9IEF]_EF4V8SP`^B#S%F6V999BO!9EEF M&]%F.Q8H`'4%9CL&)`!S"6:C)`!FB18H`&:#+C@``76C9J$D`&8K!BP`9HL6 M*`!F&Q8P`/MH!@!F4.@E[X/$!F@&`.C5[H/$`FCR`.C,[H/$`OIFQP8D`/__ M__]FQP8H`/____]FQP8L`/____]FQP8P`/____]FQP8T`&0```!FQP8X`&0` M``!F4V8SP`^B9C/`#Z)F,\`/HF9;D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0 M9E-F,\`/H@\Q9EMF4F909E-F,\`/HF9;9E-F,\`/H@\Q9EMF668KP6999AO1 M9CL6,`!U!68[!BP`<PEFHRP`9HD6,`!F(a)RXX``%UM&;'!C@`9````)"0D)"0 MD)"0D)"0D)"0D)"0D)"0D)"0D)"0D&939C/`#Z(/,69;9E)F4&939C/`#Z)F M6V939C/`#Z(/,69;9EEF*\%F668;T68[%B@`=05F.P8D`',)9J,D`&:)%B@` M9H,N.``!=;1FH20`9BL&+`!FBQ8H`&8;%C``^V@&`&90Z-;M@\0&:`8`Z(;M M@\0":`(!Z'WM@\0"^F;'!B0`_____V;'!B@`_____V;'!BP`_____V;'!C`` M_____V;'!C0`9````&;'!C@`9````&939C/`#Z)F,\`/HF8SP`^B9EN0D)"0 MD)"0D)"0D)"0D)"0D)"0D)"0D)"09E-F,\`/H@\Q9EMF4F909E-F,\`/HF9; M9E-F,\`/H@\Q9EMF668KP6999AO19CL6,`!U!68[!BP`<PEFHRP`9HD6,`!F M(a)RXX``%UM&;'!C@`9````)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D&93 M9C/`#Z(/,69;9E)F4&939C/`#Z)F6V:XT/L!`&:YH/<#`&:99O?Y9E-F,\`/ MH@\Q9EMF668KP6999AO19CL6*`!U!68[!B0`<PEFHR0`9HD6*`!F(a)RXX``%U MHV:A)`!F*P8L`&:+%B@`9AL6,`#[:`8`9E#H=>R#Q`9H!@#H)>R#Q`)H$P'H M'.R#Q`+Z9L<&)`#_____9L<&*`#_____9L<&+`#_____9L<&,`#_____9L<& M-`!D````9L<&.`!D````9E-F,\`/HF8SP`^B9C/`#Z)F6Y"0D)"0D)"0D)"0 MD)"0D)"0D)"0D)"0D&939C/`#Z(/,69;9E)F4&939C/`#Z)F6V939C/`#Z(/ M,69;9EEF*\%F668;T68[%C``=05F.P8L`',)9J,L`&:)%C``9H,N.``!=;1F MQP8X`&0```"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)!F4V8SP`^B#S%F M6V929E!F4V8SP`^B9EMF4V8SP`^B#S%F6V999BO!9EEF&]%F.Q8H`'4%9CL& M)`!S"6:C)`!FB18H`&:#+C@``76T9J$D`&8K!BP`9HL6*`!F&Q8P`/MH!@!F M4.(a)FZX/$!F@&`.C6ZH/$`F@C`>C-ZH/$`OIFQP8D`/____]FQP8H`/____]F MQP8L`/____]FQP8P`/____]FQP8T`&0```!FQP8X`&0```!F4V8SP`^B9C/` M#Z)F,\`/HF9;D)"0D)"0D)"0D)"0D)"0D)"0D)"0D)"0D&939C/`#Z(/,69; M9E)F4&939C/`#Z)F6V939C/`#Z(/,69;9EEF*\%F668;T68[%C``=05F.P8L M`',)9J,L`&:)%C``9H,N.``!=;1FQP8X`&0```"0D)"0D)"0D)"0D)"0D)"0 MD)"0D)"0D)"0D)!F4V8SP`^B#S%F6V929E!F4V8SP`^B9EMFN-#[`0!FN:#W M`P!FF6;W^6939C/`#Z(/,69;9EEF*\%F668;T68[%B@`=05F.P8D`',)9J,D M`&:)%B@`9H,N.``!=:-FH20`9BL&+`!FBQ8H`&8;%C``^V@&`&90Z,7I@\0& M:`8`Z'7I@\0":#0!Z&SI@\0"Z(CIM$S-(0`````````````````````````` M```````````````````````````````````````````````@8WEC;&5S+"!E M;7!T>0T*`"!C>6-L97,L('1E<W0-"@`@8WEC;&5S+"!E;7!T>0T*`"!C>6-L M97,L('1E<W0-"@`@8WEC;&5S+"!E;7!T>0T*`"!C>6-L97,L('1E<W0-"@`@ M8WEC;&5S+"!E;7!T>0T*`"!C>6-L97,L('1E<W0-"@`@8WEC;&5S+"!E;7!T M>0T*`"!C>6-L97,L('1E<W0-"@`@8WEC;&5S+"!E;7!T>0T*`"!C>6-L97,L M('1E<W0-"@`@8WEC;&5S+"!E;7!T>0T*`"!C>6-L97,L('1E<W0-"@`@8WEC =;&5S+"!E;7!T>0T*`"!C>6-L97,L('1E<W0-"@!L ` end ; timeit.asm Masm code 55 cycles under cmd ; ..model small, c ..686 ..stack include dosctr.asm ..data ..code codealign 8 CtBits proc uses bx dx arg mov dx,arg xor ax,ax mov bx,ax @@: shr dx,1 adc ax,bx and dx,dx jne @B ret CtBits endp codealign 8 CBits PROC push bx xor ax,ax ;XOR always clears the carry flag push dx mov bx,ax CBits0: adc ax,bx shr dx,1 ;the zero flag is set if DX = 0 jnz CBits0 ;JNZ does not look at the carry flag pop dx adc ax,bx ;don't forget the last carry bit pop bx ret CBits ENDP ..startup invoke sleep, 50 REPEAT 8 ctr_begin 100 ctr_end print dword$(eax)," cycles, empty",13,10 ctr_begin 100 ; 100 is number of reps mov eax,130000d ; something slow 1 clock mov ecx,260000d ; 26 decimal 1 clock cdq ; 1 clock idiv ecx ; divide eax by ecx 43 clocks ctr_end print dword$(eax)," cycles, test",13,10 ENDM call waitkey ..exit end |