Re: Performance overhead of get_cycles_sync

[Dor Laor <dor.laor-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]

Ingo Molnar wrote:
> * Dor Laor <dor.laor-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
>   
>> Here [include/asm-x86/tsc.h]:
>>
>> /* Like get_cycles, but make sure the CPU is synchronized. */
>> static __always_inline cycles_t get_cycles_sync(void)
>> {
>>    unsigned long long ret;
>>    unsigned eax, edx;
>>
>>    /*
>>       * Use RDTSCP if possible; it is guaranteed to be synchronous
>>      * and doesn't cause a VMEXIT on Hypervisors
>>     */
>>    alternative_io(ASM_NOP3, ".byte 0x0f,0x01,0xf9", X86_FEATURE_RDTSCP,
>>               ASM_OUTPUT2("=a" (eax), "=d" (edx)),
>>               "a" (0U), "d" (0U) : "ecx", "memory");
>>    ret = (((unsigned long long)edx) << 32) | ((unsigned long long)eax);
>>    if (ret)
>>        return ret;
>>
>>    /*
>>     * Don't do an additional sync on CPUs where we know
>>     * RDTSC is already synchronous:
>>     */
>> //    alternative_io("cpuid", ASM_NOP2, X86_FEATURE_SYNC_RDTSC,
>> //              "=a" (eax), "0" (1) : "ebx","ecx","edx","memory");
>>    rdtscll(ret);
>>     
>
> The patch below should resolve this - could you please test and Ack it? 
>   
It works, actually I already commented it out.

Acked-by: Dor Laor <dor.laor-atKUWr5tajBWk0Htik3J/w@public.gmane.org>

But this CPUID was present in v2.6.23 too, so why did it only show up in
> 2.6.24-rc for you?
>
>   
I tried to figure out but all the code movements for i386 go in the way.
In the previous email I reported to Andi that Fedora kernel 2.6.23-8 did 
not suffer from it.
Thanks for the ultra fast reply :)
Dor
> 	Ingo
>
> -------------->
> Subject: x86: fix get_cycles_sync() overhead
> From: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>
>
> get_cycles_sync() is causing massive overhead in KVM networking:
>
>    http://lkml.org/lkml/2007/12/11/54
>
> remove the explicit CPUID serialization - it causes VM exits and is
> pointless: we care about GTOD coherency but that goes to user-space
> via a syscall, and syscalls are serialization points anyway.
>
> Signed-off-by: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>
> Signed-off-by: Thomas Gleixner <tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>
> ---
>  include/asm-x86/tsc.h |   12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
>
> Index: linux-x86.q/include/asm-x86/tsc.h
> ===================================================================
> --- linux-x86.q.orig/include/asm-x86/tsc.h
> +++ linux-x86.q/include/asm-x86/tsc.h
> @@ -39,8 +39,8 @@ static __always_inline cycles_t get_cycl
>  	unsigned eax, edx;
>  
>  	/*
> -  	 * Use RDTSCP if possible; it is guaranteed to be synchronous
> - 	 * and doesn't cause a VMEXIT on Hypervisors
> +	 * Use RDTSCP if possible; it is guaranteed to be synchronous
> +	 * and doesn't cause a VMEXIT on Hypervisors
>  	 */
>  	alternative_io(ASM_NOP3, ".byte 0x0f,0x01,0xf9", X86_FEATURE_RDTSCP,
>  		       ASM_OUTPUT2("=a" (eax), "=d" (edx)),
> @@ -50,11 +50,11 @@ static __always_inline cycles_t get_cycl
>  		return ret;
>  
>  	/*
> -	 * Don't do an additional sync on CPUs where we know
> -	 * RDTSC is already synchronous:
> +	 * Use RDTSC on other CPUs. This might not be fully synchronous,
> +	 * but it's not a problem: the only coherency we care about is
> +	 * the GTOD output to user-space, and syscalls are synchronization
> +	 * points anyway:
>  	 */
> -	alternative_io("cpuid", ASM_NOP2, X86_FEATURE_SYNC_RDTSC,
> -			  "=a" (eax), "0" (1) : "ebx","ecx","edx","memory");
>  	rdtscll(ret);
>  
>  	return ret;
>
>   


-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php