xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Ian Campbell <Ian.Campbell@citrix.com>
Cc: "xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [Xen-devel] [PATCH 2/4] xen/perf: Define .glob for the different hypercalls.
Date: Tue, 3 Jul 2012 13:44:10 -0400	[thread overview]
Message-ID: <20120703174410.GC24259@phenom.dumpdata.com> (raw)
In-Reply-To: <1341332224.25743.4.camel@zakaz.uk.xensource.com>

On Tue, Jul 03, 2012 at 05:17:04PM +0100, Ian Campbell wrote:
> On Tue, 2012-07-03 at 16:40 +0100, Konrad Rzeszutek Wilk wrote:
> > This allows us in perf to have this:
> > 
> >  99.67%  [kernel]             [k] xen_hypercall_sched_op
> >   0.11%  [kernel]             [k] xen_hypercall_xen_version
> > 
> > instead of the borring ever-encompassing:
> > 
> >  99.13%  [kernel]              [k] hypercall_page
> > 
> > Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> 
> I don't suppose it would be possible to automate this somehow would it?
> Perhaps a small perl snippet consuming the output of 
>         grep ^\#define.__HYPERVISOR_ include/xen/interface/xen.h
> ?

Another way would be to do:

define NEXT_HYPERCALL(name) \
ENTRY(xen_hypercall_#name) \
.skip 32

And then
NEXT_HYPERCALL(set_trap_table);
NEXT_HYPERCALL(mmu_update);

.. and so on. And also use the .balign PAGE_SIZE at the end to pad it.


With the grep thingy I think we would also hit the large space between hypercall
#38 and #48 - which has no calls.

> 
> > ---
> >  arch/x86/xen/xen-head.S |  102 ++++++++++++++++++++++++++++++++++++++++++++++-
> >  1 files changed, 100 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S
> > index aaa7291..f6ba51d 100644
> > --- a/arch/x86/xen/xen-head.S
> > +++ b/arch/x86/xen/xen-head.S
> > @@ -28,9 +28,107 @@ ENTRY(startup_xen)
> >  	__FINIT
> >  
> >  .pushsection .text
> > -	.align PAGE_SIZE
> > +	.balign PAGE_SIZE
> >  ENTRY(hypercall_page)
> > -	.skip PAGE_SIZE
> > +ENTRY(xen_hypercall_set_trap_table)
> > +	.skip 32
> > +ENTRY(xen_hypercall_mmu_update)
> > +	.skip 32
> > +ENTRY(xen_hypercall_set_gdt)
> > +	.skip 32
> > +ENTRY(xen_hypercall_stack_switch)
> > +	.skip 32
> > +ENTRY(xen_hypercall_set_callbacks)
> > +	.skip 32
> > +ENTRY(xen_hypercall_fpu_taskswitch)
> > +	.skip 32
> > +ENTRY(xen_hypercall_sched_op_compat)
> > +	.skip 32
> > +ENTRY(xen_hypercall_platform_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_set_debugreg)
> > +	.skip 32
> > +ENTRY(xen_hypercall_get_debugreg)
> > +	.skip 32
> > +ENTRY(xen_hypercall_update_descriptor)
> > +	.skip 32
> > +ENTRY(xen_hypercall_ni)
> > +	.skip 32
> > +ENTRY(xen_hypercall_memory_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_multicall)
> > +	.skip 32
> > +ENTRY(xen_hypercall_update_va_mapping)
> > +	.skip 32
> > +ENTRY(xen_hypercall_set_timer_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_event_channel_op_compat)
> > +	.skip 32
> > +ENTRY(xen_hypercall_xen_version)
> > +	.skip 32
> > +ENTRY(xen_hypercall_console_io)
> > +	.skip 32
> > +ENTRY(xen_hypercall_physdev_op_compat)
> > +	.skip 32
> > +ENTRY(xen_hypercall_grant_table_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_vm_assist)
> > +	.skip 32
> > +ENTRY(xen_hypercall_update_va_mapping_otherdomain)
> > +	.skip 32
> > +ENTRY(xen_hypercall_iret)
> > +	.skip 32
> > +ENTRY(xen_hypercall_vcpu_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_set_segment_base)
> > +	.skip 32
> > +ENTRY(xen_hypercall_mmuext_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_xsm_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_nmi_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_sched_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_callback_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_xenoprof_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_event_channel_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_physdev_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_hvm_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_sysctl)
> > +	.skip 32
> > +ENTRY(xen_hypercall_domctl)
> > +	.skip 32
> > +ENTRY(xen_hypercall_kexec_op)
> > +	.skip 32
> > +ENTRY(xen_hypercall_tmem_op) /* 38 */
> > +	.skip 32
> > +ENTRY(xen_hypercall_rsvr)
> > +	.skip 320
> > +ENTRY(xen_hypercall_mca) /* 48 */
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_1)
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_2)
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_3)
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_4)
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_5)
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_6)
> > +	.skip 32
> > +ENTRY(xen_hypercall_arch_7)
> > +	.skip 32
> > +ENTRY(xen_hypercall_other)
> > +	.skip 2272
> > +
> >  .popsection
> >  
> >  	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS,       .asciz "linux")
> 

  reply	other threads:[~2012-07-03 17:44 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-03 15:40 [PATCH] various Xen fixes for v3.6 (v1) Konrad Rzeszutek Wilk
2012-07-03 15:40 ` [PATCH 1/4] xen/p2m: Optimize the get_phys_to_machine function a bit Konrad Rzeszutek Wilk
2012-07-09 15:41   ` [Xen-devel] " David Vrabel
2012-07-09 16:14     ` Konrad Rzeszutek Wilk
2012-07-03 15:40 ` [PATCH 2/4] xen/perf: Define .glob for the different hypercalls Konrad Rzeszutek Wilk
2012-07-03 16:07   ` Jan Beulich
2012-07-03 16:14   ` [Xen-devel] " David Vrabel
2012-07-03 16:17   ` Ian Campbell
2012-07-03 17:44     ` Konrad Rzeszutek Wilk [this message]
     [not found]   ` <4FF334E9020000780008D610@nat28.tlf.novell.com>
2012-07-03 17:38     ` Konrad Rzeszutek Wilk
2012-07-03 15:40 ` [PATCH 3/4] xen/acpi: Fix potential memory leak Konrad Rzeszutek Wilk
2012-07-03 15:40 ` [PATCH 4/4] xen/hvc: Fix up checks when the info is allocated Konrad Rzeszutek Wilk
2012-07-03 16:11   ` Jan Beulich
     [not found]   ` <4FF335C6020000780008D613@nat28.tlf.novell.com>
2012-07-03 17:36     ` Konrad Rzeszutek Wilk

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=20120703174410.GC24259@phenom.dumpdata.com \
    --to=konrad.wilk@oracle.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=xen-devel@lists.xensource.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).