Two Instructions in One Atomic Operation

[dedndave]

any way you look at it...
PUSHFD+CALL far+IRET shouldn't be that bad - lol
you are using the indirect form of CALL - that would account for 2 or 3 clock cycles - not many
ohh - i bet i know what it is - the CPU has to check IRET for privilege level changes
still - no preservation of EBX is required - you could add PUSH EBX/POP EBX into your stream
it is not the absolute number of clocks, but the variations - i can't get any believable numbers out of my machine to play with it

EDIT
i added push ebx/pop ebx - the two methods are roughly the same - and they both jump around on my machine too - lol
dang - i need a way to benchmark code

[MichaelW]

There appears to be a lot of variation between processors in the cycle counts for the two instruction sequences. This is for an old AMD K5:

Code Select Expand


14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
14 cycles, xor eax,eax | cpuid
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd
30 cycles, pushfd | call FWORD PTR | iretd


[dedndave]

please run this for me Michael - i would like to see some real numbers - lol
i can't believe anything my machine tells me


[attachment deleted by admin]

[MichaelW]

P3:

Code Select Expand


80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
80 cycles, push ebx | xor eax,eax | cpuid | pop ebx
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd
90 cycles, pushfd | call (far) | iretd

K5:

Code Select Expand


14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
14 cycles, push ebx | xor eax,eax | cpuid | pop ebx
32 cycles, pushfd | call (far) | iretd
29 cycles, pushfd | call (far) | iretd
32 cycles, pushfd | call (far) | iretd
29 cycles, pushfd | call (far) | iretd
32 cycles, pushfd | call (far) | iretd
29 cycles, pushfd | call (far) | iretd
32 cycles, pushfd | call (far) | iretd
29 cycles, pushfd | call (far) | iretd
32 cycles, pushfd | call (far) | iretd
29 cycles, pushfd | call (far) | iretd


[dedndave]

thanks Michael
i guess we can put iret to rest, then - lol
it sounded so good - had to try

[FORTRANS]

Hi,

   What if you do something like:

pushfd | call (far) | pushfd | call (far) | iretd | work | iretd

to keep the call far timings out of the loop?

Pondering,

Steve N.

[dedndave]

it is the IRET instruction that sucks up the clock cycles, i think
that is because the CPU has to check the privilege level of the return-to segment for exceptions
IRET needs an offset, segment, and flags on the stack - pushfd | call far is the fastest way to put them there
well - we could revector an interrupt and use INT nn, but that isn't going to be much faster and requires ring 0, i think
i am a bit of a n00b, so i am not sure about revectoring interrupts in protected mode windows - lol
might be able to use INT3 - that is special

[MichaelW]

pushfd | call (far) | pushfd | call (far) | iretd | work | iretd

The IRETDs would execute in the order called, even if they were physically separate.

[FORTRANS]

it is the IRET instruction that sucks up the clock cycles, i think
that is because the CPU has to check the privilege level of the return-to segment for exceptions

   Well that sorta was the question, does the CPU check the CALL segment/selector
in the same way?

IRET needs an offset, segment, and flags on the stack - pushfd | call far is the fastest way to put them there

   Or just push them and avoid the calls altogether.

well - we could revector an interrupt and use INT nn, but that isn't going to be much faster and requires ring 0, i think
i am a bit of a n00b, so i am not sure about revectoring interrupts in protected mode windows - lol
might be able to use INT3 - that is special

   Sounds like too much that way, unless the is some default set up.  (I don't
think so.)

Regards,

Steve N.

[dedndave]

Or just push them and avoid the calls altogether.

that would be pushfd | push cs | push offset | jmp
INT3 is probably the best utilization of IRET
but, i doubt it is worth the effort because the CPUID method would still be faster by a few cycles

[FORTRANS]

that would be pushfd | push cs | push offset | jmp

   I would think

pushfd | push cs | push offset | iretd

to save a jump.

Steve N.

[dedndave]

i see what you mean - that might actually be close to the cpuid method for time
i think it will be very close, but cpuid wins by 1 or 2 clocks - lol

        pushfd
        push    cs
        push    LabelA
        iret
LabelA: rdtsc

[dedndave]

here - try it out - i get 9 clocks difference - better by 1 clock cycle
i have to calculate that because of my oddball cpu - lol
it would be good to see some real numbers posted

[attachment deleted by admin]

[FORTRANS]

Hi,

   For an AMD processor.

F:\>time6
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
45 cycles, push ebx | xor eax,eax | cpuid | pop ebx
103 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
104 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
103 cycles, pushfd | push cs | push offset | iretd
104 cycles, pushfd | push cs | push offset | iretd
Press any key to exit...

   Oh, well.  I'm assuming that's from something like?

Code Select Expand


        pushfd
        push    cs
        push    LabelB
        pushfd
        push    cs
        push    LabelA
        iret
LabelA: rdtsc
        mov     ebx,eax
        iret
LabelB: rdtsc

Regards,

Steve N.

Edit:  Had the labels wrong...

[dedndave]

Thanks, Steve
it is this sequence repeated several times:

        pushfd
        push cs
        push $+3
        db 0cfh ;iret

which is equivalent to:

        pushfd
        push cs
        push LabelX
        iret
LabelX:

it does not include the RDTSC inst, as that is common to both methods
the other one is:

        push ebx
        xor  eax,eax
        cpuid
        pop  ebx

i have a couple ideas to speed it up, but the same ideas could be applied to CPUID, as well - lol