kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Avi Kivity <avi@redhat.com>
To: Marcelo Tosatti <mtosatti@redhat.com>
Cc: kvm@vger.kernel.org
Subject: Re: [patch 1/2] test: add wait parameter to on_cpu
Date: Tue, 29 Sep 2009 10:32:51 +0200	[thread overview]
Message-ID: <4AC1C633.2070403@redhat.com> (raw)
In-Reply-To: <20090928112247.375362302@redhat.com>

On 09/28/2009 01:22 PM, Marcelo Tosatti wrote:
> To determine whether to wait for IPI to finish on remote cpu.
>
> Signed-off-by: Marcelo Tosatti<mtosatti@redhat.com>
>
> Index: qemu-kvm/kvm/user/test/lib/x86/smp.c
> ===================================================================
> --- qemu-kvm.orig/kvm/user/test/lib/x86/smp.c
> +++ qemu-kvm/kvm/user/test/lib/x86/smp.c
> @@ -82,7 +82,7 @@ static void setup_smp_id(void *data)
>       asm ("mov %0, %%gs:0" : : "r"(apic_id()) : "memory");
>   }
>
> -void on_cpu(int cpu, void (*function)(void *data), void *data)
> +void on_cpu(int cpu, void (*function)(void *data), void *data, int wait)
>   {
>       spin_lock(&ipi_lock);
>       if (cpu == smp_id())
> @@ -93,9 +93,11 @@ void on_cpu(int cpu, void (*function)(vo
>   	apic_icr_write(APIC_INT_ASSERT | APIC_DEST_PHYSICAL | APIC_DM_FIXED
>                          | IPI_VECTOR,
>                          cpu);
> -	while (!ipi_done)
> -	    ;
> -	ipi_done = 0;
> +	if (wait) {
> +		while (!ipi_done)
> +		    ;
> +		ipi_done = 0;
> +	}
>       }
>       spin_unlock(&ipi_lock);
>   }
> @@ -109,6 +111,6 @@ void smp_init(void)
>
>       setup_smp_id(0);
>       for (i = 1; i<  cpu_count(); ++i)
> -        on_cpu(i, setup_smp_id, 0);
> +        on_cpu(i, setup_smp_id, 0, 1);
>
>   }
> Index: qemu-kvm/kvm/user/test/lib/x86/smp.h
> ===================================================================
> --- qemu-kvm.orig/kvm/user/test/lib/x86/smp.h
> +++ qemu-kvm/kvm/user/test/lib/x86/smp.h
> @@ -9,7 +9,7 @@ void smp_init(void);
>
>   int cpu_count(void);
>   int smp_id(void);
> -void on_cpu(int cpu, void (*function)(void *data), void *data);
> +void on_cpu(int cpu, void (*function)(void *data), void *data, int wait);
>   void spin_lock(struct spinlock *lock);
>   void spin_unlock(struct spinlock *lock);
>
> Index: qemu-kvm/kvm/user/test/x86/smptest.c
> ===================================================================
> --- qemu-kvm.orig/kvm/user/test/x86/smptest.c
> +++ qemu-kvm/kvm/user/test/x86/smptest.c
> @@ -20,6 +20,6 @@ int main()
>       ncpus = cpu_count();
>       printf("found %d cpus\n", ncpus);
>       for (i = 0; i<  ncpus; ++i)
> -	on_cpu(i, ipi_test, (void *)(long)i);
> +	on_cpu(i, ipi_test, (void *)(long)i, 1);
>       return 0;
>   }
> Index: qemu-kvm/kvm/user/test/x86/vmexit.c
> ===================================================================
> --- qemu-kvm.orig/kvm/user/test/x86/vmexit.c
> +++ qemu-kvm/kvm/user/test/x86/vmexit.c
> @@ -63,14 +63,14 @@ static void nop(void *junk)
>
>   static void ipi(void)
>   {
> -	on_cpu(1, nop, 0);
> +	on_cpu(1, nop, 0, 1);
>   }
>
>   static void ipi_halt(void)
>   {
>   	unsigned long long t;
>
> -	on_cpu(1, nop, 0);
> +	on_cpu(1, nop, 0, 1);
>   	t = rdtsc() + 2000;
>   	while (rdtsc()<  t)
>   		;
>
>    

Boolean parameters are unreadable at the call site.  I'd much prefer a 
new API (on_cpu_async(), with on_cpu_join() to wait for completion).

-- 
error compiling committee.c: too many arguments to function


  reply	other threads:[~2009-09-29  8:32 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-28 11:22 [patch 0/2] test: parallel vmexit Marcelo Tosatti
2009-09-28 11:22 ` [patch 1/2] test: add wait parameter to on_cpu Marcelo Tosatti
2009-09-29  8:32   ` Avi Kivity [this message]
2009-09-28 11:22 ` [patch 2/2] test: vmexit: run parallel tests on each cpu Marcelo Tosatti
2009-09-29  8:35   ` Avi Kivity
2009-10-05 13:07 ` [patch 0/2] kvm: test: parallel vmexit (v2) Marcelo Tosatti
2009-10-05 13:07   ` [patch 1/2] test: add on_cpu_async Marcelo Tosatti
2009-10-05 13:07   ` [patch 2/2] test: vmexit: run parallel tests on each cpu Marcelo Tosatti
2009-10-06  0:56     ` [patch 3/3] test: vmexit: inl from pmtimer Marcelo Tosatti
2009-10-06  9:56   ` [patch 0/2] kvm: test: parallel vmexit (v2) Avi Kivity

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4AC1C633.2070403@redhat.com \
    --to=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).