vcpu is blocked by other domain.
3)Current consider_steal_time() handles stolen and blocked tick.
Other running tick is considered on timer_interrupt.
Then I reply your questions in line
Isaku Yamahata [EMAIL PROTECTED] wrote:
On Thu, May 15, 2008 at 06:02:00PM +0900, Atsushi SAKAI wrote
in it.
As you suggested, if we consider the case of manu vcpu on one pcpu,
the alpha value expected to become large value.
But no problem in it.
Thanks
Atsushi SAKAI
Isaku Yamahata [EMAIL PROTECTED] wrote:
On Wed, May 14, 2008 at 06:24:07PM +0900, Atsushi SAKAI wrote:
local_cpu_data
time account TICK is different (1/HZ).
So In some cases, stolentick = blocked + stolen (in TICK)
Thanks
Atsushi SAKAI
Isaku Yamahata [EMAIL PROTECTED] wrote:
On Tue, May 13, 2008 at 08:24:33PM +0900, Atsushi SAKAI wrote:
The stolentick is composed of blocked, stolen and truely running time
intention with stolentick, stolen and blocked?
This stolentick means same as delta_cpu in x86 code.
other variables are same as that of x86.
Is this answering your question?
Thanks
Atsushi SAKAI
___
Xen-ia64-devel mailing list
Xen-ia64-devel
considered on stolentick++;
Thanks
Atsushi SAKAI
Isaku Yamahata [EMAIL PROTECTED] wrote:
On Fri, May 09, 2008 at 03:48:24PM +0900, Atsushi SAKAI wrote:
Isaku Yamahata [EMAIL PROTECTED] wrote:
do_div(stolentick, NS_PER_TICK);
stolentick++;
do_div
the timer interrupt handler was invoked.
Thanks
Atsushi SAKAI
Isaku Yamahata [EMAIL PROTECTED] wrote:
stolentick is used to clip sotlen and blocked.
It looks like it should be the diff of now and the tick when
the timer interrupt handler was invoked, not new_itm.
Atsushi, any comment
Hi,
This is clean up patch for linux-defconfig(xen, xen0, xenU)
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
Thanks
Atsushi SAKAI
Alex Williamson [EMAIL PROTECTED] wrote:
On Tue, 2007-10-16 at 11:06 +0900, Atsushi SAKAI wrote:
Hi, Alex
Thank you for your comments.
I added
Hi,
This patch intends to modify
PV/buildconfigs/linux-defconfig_xen_ia64
to default setting(from PV/.config).
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
Thanks
Atsushi SAKAI
modify_buildconfig_ia64.patch
Description: Binary data
___
Xen-ia64
Hi, Alex
Thank you for your comments.
I added xen0 and xenU patch.
Thanks
Atsushi SAKAI
Alex Williamson [EMAIL PROTECTED] wrote:
On Mon, 2007-10-15 at 22:44 +0900, Atsushi SAKAI wrote:
Hi,
This patch intends to modify
PV/buildconfigs/linux-defconfig_xen_ia64
to default setting
this work at this moment.)
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
From Keir notice, Xen 3.2.0 is very close to code freeze,
So I posted this current version patch.
(If someone plan to implement sync feature within 2-weeks,
it should be appreciated.)
Thanks
Atsushi SAKAI
Hi,
I compiled xen-ia64-unstable CS:14419 and works fine.
of course libxc works fine.
Which version you are using?
Thanks
Atsushi SAKAI
Rodrigo Lord [EMAIL PROTECTED] wrote:
Hi!
I was trying to compile xen-ia64-unstable on a Itanium2 (HP rx2600 with 2
processors) and an error about
Hi, Alex
Thank you for commiting patches.
And thanks for fixing build failure by merging xen-unstable.
This is the last one of patches(3rd-patch).
(based on xen-ia64-unstable 14297)
As you suggested,
I unified from stolen and blocked loop to one.
Thanks
Atsushi SAKAI
it as my backlog.
Thanks
Atsushi SAKAI
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http://lists.xensource.com/xen-ia64-devel
This is PVdomain side patch.
Thanks
Atsushi SAKAI
steal_time_pvdom.patch
Description: Binary data
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http://lists.xensource.com/xen-ia64-devel
Hi, Dodrigo
Credit scheduler just do weight based balancing in principle.
Is this your answer?
Anyway What do you mean vcpu balancing?
But this kind of discussion should be done in xen-devel.
Thanks
Atsushi SAKAI
Rodrigo Lord [EMAIL PROTECTED] wrote:
Hi!
I would like to know if the SMP
Hi, Isaku
Thank you for your suggestion.
Do you have any idea to consolidate steal time for various arch?
Thanks
Atsushi SAKAI
Isaku Yamahata [EMAIL PROTECTED] wrote:
On Wed, Feb 07, 2007 at 05:55:23PM +0900, Atsushi SAKAI wrote:
This patch intends to consider steal time accounting
Hi,
This patch adds localtime setting for PV/IA64 domain.
This will work if xen-unstable 13835 patch is applied.
http://xenbits2.xensource.com/xen-unstable.hg?rev/d609929f1408
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
Any comments are appreciated.
Thanks
Atsushi SAKAI
patch of steal time.
Thanks
Atsushi SAKAI
consider_steal_time.patch
Description: Binary data
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http://lists.xensource.com/xen-ia64-devel
import linux/arch/ia64/kernel/time.c
Thanks
Atsushi SAKAI
import_time_c.patch
Description: Binary data
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http://lists.xensource.com/xen-ia64-devel
.
Anyway,
shared_info uses MAX_VIRT_CPUS array.
Is this permitted to use MAX_VIRT_CPUS array in arch_shared_info?
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
Thanks
Atsushi SAKAI
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
Hi, Tang.
Let me clearify.
Are you testing domVTI or DomU?
If domVTI case, you need to add Paradriver(PV-on-HVM),
This is not included in RHEL4u4.
You need to make it.
Thanks
Atsushi SAKAI
Tang, Wei [EMAIL PROTECTED] wrote:
Sakai:
Thanks for your email, my VM user normal Linux RHEL
Hi, Wei
It should be run on both.
Thanks
Atsushi SAKAI
Tang, Wei [EMAIL PROTECTED] wrote:
Hi, All:
Does anyone can tell me which release version can support VNIF? 3.0.3 or
3.0.4?
Thanks
Wei
Intel Xen Virtualization Enabling
Global Developer Relations Division
フタ
Hi, Alex and Aron
Thank you for your various comments,
I attach the patche which reflect this discussion.
Please edit the comment line in patches, as you like.
I change the last line of document from previous mail.
Thanks
Atsushi SAKAI
==
This patch
at update_last_itc.
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
Thanks
Atsushi SAKAI
fix-vti-oops.patch
Description: Binary data
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http://lists.xensource.com/xen-ia64-devel
Hi, Dietmar
As you suggested.
Currently HYPERVISOR_set_timer_op is not used.
for more detail, please see following mail.
http://lists.xensource.com/archives/html/xen-ia64-devel/2006-07/msg00056.html
Thanks
Atsushi SAKAI
Dietmar Hahn write on 2007年1月15日 23:01:
Hi,
Hi Dietmar,
The attached
Hi, Markus
I tested on SMP DomU at xen-ia64-unstable 13083.
It seems works fine.
Thanks
Atsushi SAKAI
xenfb_update_screen() calls zap_page_range() while holding spinlock
mm_lock. Big no-no.
Changeset 13018:c98ca86138a7422cdf9b15d87c95619b7277bb6a merely sweeps
the bug under the carpet
need to use spin_lock_irqsave in xenfb_update()?
I understand the locking but why irqsave is required at this point?
Thanks
Atsushi SAKAI
Atsushi SAKAI [EMAIL PROTECTED] writes:
Hi, Markus
I am checking VFB with SMP
with xen-ia64-unstable cset 12886.
But It seems strange behavior.
(call
Hi,
I do simple test xen-3.0.4-testing on IA64.
(CSet12901)
SMP DomU boot successfully
VFB w/ UP DomU works fine.
I am just using Madison machine at this moment.
So I cannot test DomVTI.
Thanks
Atsushi SAKAI
___
Xen-ia64-devel mailing list
Xen
Hi, Markus
I am checking VFB with SMP
with xen-ia64-unstable cset 12886.
But It seems strange behavior.
(call trace appeared from DomU xm console output)
Did you test it on SMP DomU?
Of course UP DomU seems fine at this moment.
Thanks
Atsushi SAKAI
Hi、Wing
Current change set (for example xen-ia64-unstable 12795)
does not run xentrace.
Need to wait Isaku-patch[5/5] applied.
Or back to older change sets.
If this is fixed in xen-ia64-unstable, I will report you.
Thanks
Atsushi SAKAI
Hi Hiroya:
Thanks for your guide. But I meet
Hi, Markus
Thank you
I confirmed to run on IA64.
Thanks
Atsushi SAKAI
Atsushi SAKAI [EMAIL PROTECTED] writes:
Hi, Markus
Thank you for your suggestion.
Would you please post sample vfb config file?
Or any document for latest version.
Anyway, your vfb patch makes large changes compared
in config file
seems to be vfbif = xx not vfb = xx
Thanks
Atsushi SAKAI
Atsushi SAKAI [EMAIL PROTECTED] writes:
Hi, Markus
Thank you for your suggestion.
My config is attached as follows(see below)
Anyway, I cannot find vfb and vkbd definition in RHEL5 package.
Please give me appropriate
-xenU.img
memory = 1024
name = rhel4-sakaia0
disk = [ 'file:/xen/image/fc6t3.img,hda1,w' ]
vif = [ '' ]
root = /dev/hda1 ro
extra = nomca ide0=noprobe ide1=noprobe ide2=noprobe ide3=noprobe xencons=xvc
vnc = 1
vncpasswd=' '
Thanks
Atsushi SAKAI
Atsushi SAKAI [EMAIL PROTECTED] writes:
Hi
Hi, Yu
Don't forget IA64 code!
Thanks
Atsushi SAKAI
I have a similar patch for this, which is generated in
linux-2.6.16.33-xen/.config automatically.
See attachment defbuild.patch.
Pae_new_config is used to build x86_32 with PAE enabled.
Ping
-Original Message-
From: [EMAIL
=xenfb_wait_for_state
=xs_read_watch
this point in xen-vncfb is initialize phase of vfb.
So vfb should not be appeared.
Thanks
Atsushi SAKAI
Atsushi SAKAI [EMAIL PROTECTED] writes:
Hi, Markus
This is just a status.
I tested on this.
It has still in serial console problem.
(I think this problem
Hi, Isaku
Thank you for your comments.
I am just caring ltrace problem.
I just check dom0 boot and ltrace command.
If any suggestions, please fix it.
(Since I have no idea on this.)
Thanks
Atsushi SAKAI
This patch is incomplete.
If the guest kernel issues break 0x80001 or 0x80002
the CONFIG_XEN_FRAMEBUFFER=n.
From this environment,
I guess this problem is in serial console.
Anyway xen-vncfb is running (not defunct).
Thanks
Atsushi SAKAI
Hi Atsushi,
Could you give my latest patch a try on IA-64? I didn't merge your
patch literally, so I might have broken it.
Cheers,
Markus
/hpsim.S
Do you have any idea to solve this problem?
Thanks
Atsushi SAKAI
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http://lists.xensource.com/xen-ia64-devel
Hi, Markus
I checked your fixes.
It seems fine in source code base.
I will run the test next week.
Thanks
Atsushi SAKAI
Hi Atsushi,
Could you give my latest patch a try on IA-64? I didn't merge your
patch literally, so I might have broken it.
Cheers,
Markus
-by: Isaku Yamahata [EMAIL PROTECTED]
Signed-off-by: Masami Watanabe [EMAIL PROTECTED]
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
From seeing your patches policy,
We do memory address translation in Dom0 application side.
Thanks
Atsushi SAKAI
PV framebuffer backend. Derived from http
Hi, Tristan
Thank you for your comments.
I am wondering the code of arch/i386/pci/pcifront.c and irq-xen.c for x86.
(IA64 does not have it)
The [EMAIL PROTECTED]/i386/pci/irq-xen.c is pci_sal_read/write for IA64.
(DomU/VTI cannot access PCI configuration)
Thanks
Atsushi SAKAI
Le Vendredi
I'm just watching through the source code of PCI front/back driver.
But I did not understand the interface of Para/IA64 kernel.
(x86 has driver at linux-sparce/arch/x86/driver/pci
but IA64 does not have such.)
If anyone know the interface, please let me know thanks.
Thanks,
Atsushi SAKAI
Hi, Tristan
That's fine that GRUB support IA64 platform!
Do you plan to support TrustedGRUB?
http://sourceforge.net/projects/trustedgrub/
Thanks
Atsushi SAKAI
Hi,
here is my port of grub to ia64. I hope it has all the requested features
(kernel, initrd, modules, relocation and fpswa
Hi, All
This patch intends to add VTI-related and fw_hypercall counters.
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
Signed-off-by: Hiroya INAKOSHI [EMAIL PROTECTED]
Thanks
Atsushi SAKAI
vti_perfc.patch
Description: Binary data
___
Xen
Hi, Anthony
I tested following changes, it works fine.
(13:44 Run in 2xDomU Kernel Compile
and DomVTI(nosmp) boot test)
Thanks
Atsushi SAKAI
+++ b/xen/arch/ia64/xen/hypercall.cThu Aug 24 11:48:35 2006 -0600
@@ -235,7 +235,12 @@ fw_hypercall (struct pt_regs *regs
tested your proposed configuration.
I will test it in night run today.
Thanks
Atsushi SAKAI
That's Ok for me.
@@ -122,6 +122,7 @@ void schedule_tail(struct vcpu *prev)
shared_info-vcpu_info[current-vcpu_id].evtchn_upcall_mask;
__ia64_per_cpu_var(current_psr_ic_addr
Hi, Isaku
Please tell CSet #, you are intended for ?
Our group people tried to compile your patches at TAKE4, but failed.
To skip this issue, Please tell CSet #.
Thanks
Atsushi SAKAI
Hi. I posted performance tuning patches.
It must be evaluated for its effect before commit.
(or improve them
Hi, Anthony
Could you explain following cases?
Anyway, thank you for your patch of pal_halt_light domU.
Thanks,
Atsushi SAKAI
In some scenarios, xen_timer_interrupt will not set machine itm,
this may make this LP not receive timer interrupt any more.
Signed-off-by: Anthony Xu [EMAIL
is running but timer(VIRQ_ITC) is
stopped.
(References)
Yamahata's for VTI migrate timer.
http://lists.xensource.com/archives/html/xen-ia64-devel/2006-07/msg00375.html
stop_timer
http://lists.xensource.com/archives/html/xen-ia64-devel/2006-07/msg00171.html
Signed-off-by: Atsushi SAKAI [EMAIL
this patch does not need to add vcpu_pend_timer fixes.
But this should be fixed in sometime.
(another reason)
Thanks,
Atsushi SAKAI
Hi Atsushi.
On Tue, Aug 22, 2006 at 02:44:34PM +0900, Atsushi SAKAI wrote:
As you pointed out, vcpu_pend_timer() is usually used as current.
This problem does
-itm_delta)+NOW;
-Anthony
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Atsushi
SAKAI
Sent: 2006?8?21? 19:39
To: xen-ia64-devel@lists.xensource.com
Subject: [Xen-ia64-devel] [PATCH] pal_halt_light emulatefor domU TAKE3
This is a pal_halt_light emulation
/archives/html/xen-ia64-devel/2006-07/msg00171.html
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
pal_halt_light_emulate_take3.patch
Description: Binary data
___
Xen-ia64-devel mailing list
Xen-ia64-devel@lists.xensource.com
http
Hi Anthony
Thank you for your comment,
I am already trying test in both cases(for more than 16Hour 4CPU).
But I select current by code simplicity.
I agree your point, I will fix it.
Thanks
Atsushi SAKAI
Hi SAKAI,
Good effort,
I have one small comment
+vcpu_get_next_timer_ns
to use as vcpu==current)
2)Change the order of send_guest_vcpu_virq(vcpu, VIRQ_ITC)
PSCBX(vcpu, domain_itm_last) = PSCBX(vcpu, domain_itm);
Thanks
Atsushi SAKAI
Hi Atsuhi.
On Mon, Aug 21, 2006 at 08:38:33PM +0900, Atsushi SAKAI wrote:
2)The VIRQ_ITC cycle is destroyed
in case vcpu
Hi, Tristan
Is there any plan to add a counter for vcpu migration counting?
I want to see the correlation with these patch.
Thanks,
Atsushi SAKAI
Le Mercredi 09 Ao〓 2006 13:27, Isaku Yamahata a 馗rit :
Hi all
Hi
some comments enclosed.
These patches are for performance tuning TAKE 3
Hi, Tristan
O.K.
I will add it.
It seems that xenperf is your territory.
So I need your permission:-)
Thanks,
Atsushi SAKAI
Le Mercredi 09 Ao〓 2006 13:50, Atsushi SAKAI a 馗rit :
Hi, Tristan
Is there any plan to add a counter for vcpu migration counting?
I want to see the correlation
Hi, Alex
Thank you for your action.
c.f.
I am a business trip to Japan north region(TOYAMA)
from Thursday to Friday.
It makes delay.
Thanks,
Atsushi SAKAI
On Wed, 2006-07-19 at 09:03 -0400, Alex Williamson wrote:
On Wed, 2006-07-19 at 20:16 +0900, Atsushi SAKAI wrote:
For the short term
exclude VIRQ_ITC interrupt duing that time
window.
and this problem is reproduced by using lmbench.(Thanks)
Thanks
Atsushi SAKAI
Hi, Yamahata,
If this domain depends on hlt_timer to wake it up,
The scenario you described can appear,
TIMER_SLOP can reduce this situation, but can't eliminate
Hi, Alex
I am checking it.
But I can not reproduce the oops message.
(So I cannot check the modified candidate code)
Would you give me a condition ?
Thanks
Atsushi SAKAI
The Oops message is back on SMP domUs. Is this from the
PAL_HALT_LIGHT emulation? Thanks,
Alex
--
Alex
,
Atsushi SAKAI
Sakai-san,
Another short term approach,
When emulating PAL_HALT_LIGHT
If(dom0)
Ia64_pal_halt_hight();
Else //domU
do_block;
Thanks,
anthony
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Atsushi
SAKAI
Sent: 2006?7?11? 21:46
Hi,
My primary motivation is correct display of xenmon.py and xentop in BVT/CREDIT.
(N.B. SEDF displayed as same as x86, but BVT/CREDIT are not)
If only domU emulation is applied, it is a half way from my motivation.
Is dom0 dispatch another home work?
Thanks,
Atsushi SAKAI
From: Atsushi SAKAI
, vcpu_set_next_timer is not needed in source code base.
Since I just need VIRQ_ITC event.
I will test it and submit the patch after applying your comments.
Thanks
Atsushi SAKAI
From: Atsushi SAKAI
Hi, All
This patch emulates Guest PAL_HALT_LIGHT on domU by using
do_block and timer.
It also adds
Hi, Kevin
Thank you for your comments.
I agree your points.
I will change it as your comments.
Anyway, I should change the name of function(vcpu_get_next_timer),
because the meaning is changed.:-)
Thanks,
Atsushi SAKAI
From: Atsushi SAKAI [mailto:[EMAIL PROTECTED]
Hi, Kevin
Sorry
Hi, All
This patch emulates Guest PAL_HALT_LIGHT on domU by using do_block and timer.
It also adds the function of the timer interrupt to domU at the vcpu woke up.
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
About the timer interrupt to domU
The function xen_timer_interrupt on Xen only
to emulate PAL_HALT_LIGHT on the Xen Hypervisor.
This model does not effect on VIRQ_ITC and itc/itm model.
Thanks,
Atsushi SAKAI.
Hi Sakai.
xenLinux/x86 paravirtualizes idle loop to get timer interrupt
while cpu halting.
More exactly safe_halt() is paravirtualized using
Hi, Tristan.
I have no idea without changing the linux scheduler code.
I also want to know the idea to solve your problem.
If the linux stops over 10 seconds, it appears.
On Xen, that problem is common in x86/VTx, IA64/noVTI and VTI.
Thanks,
Atsushi SAKAI
I have a question related
in x86/VTx, IA64/noVTI and VTI.
soft lockup code:
https://www.codeblog.org/gonzui/markup/linux-2.6.16.18/kernel/softlockup.c?q=package%3Alinux-2.6.16.18+fundef%3Asoftlockup_tick#l47
Thanks,
Atsushi SAKAI
I have a question related to the timer.
If you pause/unpause a domain for more than 10 secs
Hi, All
This patch emulates Guest PAL_HALT_LIGHT on domU by using do_block and timer.
It also adds the function of the timer event sending to domU at the vcpu woke
up.
Signed-off-by: Atsushi SAKAI [EMAIL PROTECTED]
About the timer event sending to domU
The function xen_timer_interrupt
(Physical)
・Gypsy Domain 1vcpu
・Blocking Domain 1 vcpu
VTI
・6CPU machine
・Xen 2CPU(Physical)
・Gypsy Domain 1vcpu
・Blocking Domain 1 vcpu
Thanks.
Reiji TAZAKI
Atsushi SAKAI
Hi
I am maybe late on this issue but has anyone tried credit scheduler on
xen/ia64 ?
I suppose it doesn't work well
Hi All
I have a question about Domain time update in IA64.
(update_vcpu_system_time)
In x86, this function is working.
But in ia64, the function is null at this moment.
Is there any plan to change the function specification ?
Atsushi Sakai
70 matches
Mail list logo