Re: [Xen-devel] [PATCH v2 2/2] xentrace: Add TRC_HVM_VCHIP

[George Dunlap <george.dunlap@xxxxxxxxxxxxx>]

On 03/28/2014 04:56 PM, Don Slutz wrote:
> This add a set of trace events that track the setup of various
> virtual chips related to timers in domU.
>
> This set is hpet, pit (i8253, i8254), rtc (MC146818), apic (lapic),
> and pic (i8259).  The pmtimer is not traced since it does not have a
> changeable rate.
>
> Signed-off-by: Don Slutz <dslutz@xxxxxxxxxxx>
> ---
>   tools/xentrace/formats     | 18 ++++++++++++++++++
>   xen/arch/x86/hvm/hpet.c    |  5 +++++
>   xen/arch/x86/hvm/i8254.c   |  9 +++++++++
>   xen/arch/x86/hvm/rtc.c     |  9 +++++++++
>   xen/arch/x86/hvm/vlapic.c  | 16 ++++++++++++++++
>   xen/arch/x86/hvm/vpic.c    | 11 +++++++++++
>   xen/include/public/trace.h | 20 ++++++++++++++++++++
>   7 files changed, 88 insertions(+)
>
> diff --git a/tools/xentrace/formats b/tools/xentrace/formats
> index 67fd42d..1d28e2a 100644
> --- a/tools/xentrace/formats
> +++ b/tools/xentrace/formats
> @@ -152,3 +152,21 @@
>   0x00802006  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  assign_vector [ irq = %(1)d = 
> vector 0x%(2)x, CPU mask: 0x%(3)08x ]
>   0x00802007  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  bogus_vector [ 0x%(1)x ]
>   0x00802008  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  do_irq [ irq = %(1)d, began = 
> %(2)dus, ended = %(3)dus ]
> +
> +0x00084001  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  hpet create [ tn = %(1)d, irq 
> = %(2)d, delta = %(3)d, period = %(4)d ]
> +0x00084002  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  pit create [ delta = %(1)d, 
> period = %(2)d ]
> +0x00084003  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  rtc create [ delta = %(1)d, 
> period = %(2)d ]
> +0x00084004  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vlapic create [ delta = %(1)d, 
> period = %(2)d, irq = %(3)d ]
> +0x00084005  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  hpet destroy [ tn = %(1)d ]
> +0x00084006  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  pit destroy  [ ]
> +0x00084007  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  rtc destroy [ ]
> +0x00084008  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vlapic destroy [ ]
> +0x00084009  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  pit callback [ ]
> +0x0008400a  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vlapic callback [ ]
> +0x0008400b  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vpic_update_int_output [ 
> int_output = %(1)d, is_master = %(2)d, irq = %(3)d ]
> +0x0008400c  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vpic vcpu_kick [ irq = %(1)d ]
> +0x0008400d  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  __vpic_intack [ is_master = 
> %(1)d, irq = %(2)d ]
> +0x0008400e  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vpic_irq_positive_edge [ irq = 
> %(1)d ]
> +0x0008400f  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vpic_irq_negative_edge [ irq = 
> %(1)d ]
> +0x00084010  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vpic_ack_pending_irq [ 
> accept_pic_intr = %(1)d, int_output = %(2)d ]
> +0x00084011  CPU%(cpu)d  %(tsc)d (+%(reltsc)8d)  vlapic_accept_pic_intr [ 
> i8259_target = %(1)d, accept_pic_int = %(2)d ]
> diff --git a/xen/arch/x86/hvm/hpet.c b/xen/arch/x86/hvm/hpet.c
> index e1e769a..bb6de72 100644
> --- a/xen/arch/x86/hvm/hpet.c
> +++ b/xen/arch/x86/hvm/hpet.c
> @@ -24,6 +24,7 @@
>   #include <asm/hpet.h>
>   #include <xen/sched.h>
>   #include <xen/event.h>
> +#include <xen/trace.h>
>
>   #define domain_vhpet(x) (&(x)->arch.hvm_domain.pl_time.vhpet)
>   #define vcpu_vhpet(x)   (domain_vhpet((x)->domain))
> @@ -191,6 +192,7 @@ static void hpet_stop_timer(HPETState *h, unsigned int tn)
>   {
>       ASSERT(tn < HPET_TIMER_NUM);
>       ASSERT(spin_is_locked(&h->lock));
> +    TRACE_1D(TRC_HVM_VCHIP_HPET_STOP_TIMER, tn);
>       destroy_periodic_time(&h->pt[tn]);
>       /* read the comparator to get it updated so a read while stopped will
>        * return the expected value. */
> @@ -255,6 +257,9 @@ static void hpet_set_timer(HPETState *h, unsigned int tn)
>        * being enabled (now).
>        */
>       oneshot = !timer_is_periodic(h, tn);
> +    TRACE_4D(TRC_HVM_VCHIP_HPET_START_TIMER, tn, irq,
> +             hpet_tick_to_ns(h, diff),
This will only be 32 bits wide; is there ever any risk that this may 
exceed 4 seconds (4 billion nanoseconds)?

> diff --git a/xen/include/public/trace.h b/xen/include/public/trace.h
> index e2f60a6..410f0fe 100644
> --- a/xen/include/public/trace.h
> +++ b/xen/include/public/trace.h
> @@ -52,6 +52,7 @@
>   /* trace subclasses for SVM */
>   #define TRC_HVM_ENTRYEXIT 0x00081000   /* VMENTRY and #VMEXIT       */
>   #define TRC_HVM_HANDLER   0x00082000   /* various HVM handlers      */
> +#define TRC_HVM_VCHIP     0x00084000   /* virtual chips */
Maybe TRC_HVM_EMUL, for emulated devices?

 -George


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel