Xen devel - Sep 2011 - ring buffer overflow

Hi,

Does anybody know whether the ring buffer between front end and back
end will suffer from overflow? I just wonder if the ring buffer will
be full and drop some packets when the Net I/O load is very heavy.
BTW, If I want to change the size of i/o ring buffer, how should I do?
I tried to reset the NET_TX_RING_SIZE and NET_RX_RING_SIZE in both
front end and back end, but it seems doesn''t work. Thanks.

Regards,
Cong

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Fri, 2011-09-30 at 05:18 +0100, David Xu wrote:
> Hi,
> 
> Does anybody know whether the ring buffer between front end and back
> end will suffer from overflow? I just wonder if the ring buffer will
> be full and drop some packets when the Net I/O load is very heavy.
In the case of networking whichever end is putting stuff on the ring
checks that there is enough room and will stop the queue when it cannot
transmit any more and restart when room becomes available.
> BTW, If I want to change the size of i/o ring buffer, how should I do?
> I tried to reset the NET_TX_RING_SIZE and NET_RX_RING_SIZE in both
> front end and back end, but it seems doesn''t work. Thanks.
Currently the rings are limited to 1 page so if you want to increase the
size you would need to add multipage ring support to the network
protocol. There have been patches to do this for the blk protocol but I
do not recall any for the net protocol.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hi,

2011/9/29 Ian Campbell <Ian.Campbell@citrix.com>:
> On Fri, 2011-09-30 at 05:18 +0100, David Xu wrote:
>> Hi,
>>
>> Does anybody know whether the ring buffer between front end and back
>> end will suffer from overflow? I just wonder if the ring buffer will
>> be full and drop some packets when the Net I/O load is very heavy.
>
> In the case of networking whichever end is putting stuff on the ring
> checks that there is enough room and will stop the queue when it cannot
> transmit any more and restart when room becomes available.
You mean even there is not enough room in ring buffer, xen will * not
drop the packets * and just delay the transmission. I used httperf to
measure the performance of web server running in a VM (The workload in
this VM is mixed, so it can not benefit from boost mechanism. The net
i/o suffers from relatively high latency which depends on the number
of VMs in the system). I found that with the increase of request rate
in client side, the connection rate will drop and the connection time
will increase dramatically. And the retransmission appears when the
request rate is over than a quantum. So I doubted that the http/tcp
connection suffer from the packets drop when the ring buffer is full
because of high request rate.
>
>> BTW, If I want to change the size of i/o ring buffer, how should I do?
>> I tried to reset the NET_TX_RING_SIZE and NET_RX_RING_SIZE in both
>> front end and back end, but it seems doesn''t work. Thanks.
>
> Currently the rings are limited to 1 page so if you want to increase the
> size you would need to add multipage ring support to the network
> protocol. There have been patches to do this for the blk protocol but I
> do not recall any for the net protocol.
Yes, increasing the size is relatively hard. So I just want reduce the
size of ring buffer to make sure my doubt described above. I directly
set  NET_TX_RING_SIZE and  NET_RX_RING_SIZE to 128, but it doesn''t seem
to work.

Regards,
Cong

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Fri, 2011-09-30 at 15:44 +0100, David Xu wrote:
> Hi,
> 
> 2011/9/29 Ian Campbell <Ian.Campbell@citrix.com>:
> > On Fri, 2011-09-30 at 05:18 +0100, David Xu wrote:
> >> Hi,
> >>
> >> Does anybody know whether the ring buffer between front end and
back
> >> end will suffer from overflow? I just wonder if the ring buffer
will
> >> be full and drop some packets when the Net I/O load is very heavy.
> >
> > In the case of networking whichever end is putting stuff on the ring
> > checks that there is enough room and will stop the queue when it
cannot
> > transmit any more and restart when room becomes available.
> 
> You mean even there is not enough room in ring buffer, xen will * not
> drop the packets * and just delay the transmission.
It''s not Xen but rather the kernel back and front ends which is
involved
here. You can examine the hard_start_xmit functions in both netback and
netfront to determine for yourself whether or not packets can be dropped
and when.
>  I used httperf to
> measure the performance of web server running in a VM (The workload in
> this VM is mixed, so it can not benefit from boost mechanism. The net
> i/o suffers from relatively high latency which depends on the number
> of VMs in the system). I found that with the increase of request rate
> in client side, the connection rate will drop and the connection time
> will increase dramatically. And the retransmission appears when the
> request rate is over than a quantum. So I doubted that the http/tcp
> connection suffer from the packets drop when the ring buffer is full
> because of high request rate.
> 
> >
> >> BTW, If I want to change the size of i/o ring buffer, how should I
do?
> >> I tried to reset the NET_TX_RING_SIZE and NET_RX_RING_SIZE in both
> >> front end and back end, but it seems doesn''t work.
Thanks.
> >
> > Currently the rings are limited to 1 page so if you want to increase
the
> > size you would need to add multipage ring support to the network
> > protocol. There have been patches to do this for the blk protocol but
I
> > do not recall any for the net protocol.
> 
> Yes, increasing the size is relatively hard. So I just want reduce the
> size of ring buffer to make sure my doubt described above. I directly
> set  NET_TX_RING_SIZE and  NET_RX_RING_SIZE to 128, but it doesn''t
seem to work.
You need to make sure both ends of the connection agree on the ring
size.

I''m afraid this is not a very common thing to want to do so if you want
to persist with this approach I''m afraid you''ll have to do
some
debugging.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hi,

I made the experiment with httperf again, and used tcpdump to capture
the packets on both server side and client side. I found that if there
was some retransmission (e.t. client send several syn to server
because it didn''t receive ack in time), the vif only receive the last
packet (syn) and missed the former ones. So I think there are some
issues happened between the eth0(veth0) and vif in the dom0. But if I
use a low latency scheduler written by myself (only modify the client
scheduler and didn''t touch other parts ), there will not be
retransmission or there is very few retransmission. I am not familiar
with netback of xen. Can you give me some suggestion? Or which part
source code I need to check to find the reason of packets loss between
eth0(veth0) and vif in dom0? Thanks.

Regards,
Cong

2011/9/30 Ian Campbell
<Ian.Campbell@eu.citrix.com>:
> On Fri, 2011-09-30 at 15:44 +0100, David Xu wrote:
>> Hi,
>>
>> 2011/9/29 Ian Campbell <Ian.Campbell@citrix.com>:
>> > On Fri, 2011-09-30 at 05:18 +0100, David Xu wrote:
>> >> Hi,
>> >>
>> >> Does anybody know whether the ring buffer between front end
and back
>> >> end will suffer from overflow? I just wonder if the ring
buffer will
>> >> be full and drop some packets when the Net I/O load is very
heavy.
>> >
>> > In the case of networking whichever end is putting stuff on the
ring
>> > checks that there is enough room and will stop the queue when it
cannot
>> > transmit any more and restart when room becomes available.
>>
>> You mean even there is not enough room in ring buffer, xen will * not
>> drop the packets * and just delay the transmission.
>
> It''s not Xen but rather the kernel back and front ends which is
involved
> here. You can examine the hard_start_xmit functions in both netback and
> netfront to determine for yourself whether or not packets can be dropped
> and when.
>
>>  I used httperf to
>> measure the performance of web server running in a VM (The workload in
>> this VM is mixed, so it can not benefit from boost mechanism. The net
>> i/o suffers from relatively high latency which depends on the number
>> of VMs in the system). I found that with the increase of request rate
>> in client side, the connection rate will drop and the connection time
>> will increase dramatically. And the retransmission appears when the
>> request rate is over than a quantum. So I doubted that the http/tcp
>> connection suffer from the packets drop when the ring buffer is full
>> because of high request rate.
>>
>> >
>> >> BTW, If I want to change the size of i/o ring buffer, how
should I do?
>> >> I tried to reset the NET_TX_RING_SIZE and NET_RX_RING_SIZE in
both
>> >> front end and back end, but it seems doesn''t work.
Thanks.
>> >
>> > Currently the rings are limited to 1 page so if you want to
increase the
>> > size you would need to add multipage ring support to the network
>> > protocol. There have been patches to do this for the blk protocol
but I
>> > do not recall any for the net protocol.
>>
>> Yes, increasing the size is relatively hard. So I just want reduce the
>> size of ring buffer to make sure my doubt described above. I directly
>> set  NET_TX_RING_SIZE and  NET_RX_RING_SIZE to 128, but it
doesn''t seem to work.
>
> You need to make sure both ends of the connection agree on the ring
> size.
>
> I''m afraid this is not a very common thing to want to do so if you
want
> to persist with this approach I''m afraid you''ll have to
do some
> debugging.
>
> Ian.
>
>
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Please don''t top post.

On Tue, 2011-10-04 at 16:17 +0100, David Xu wrote:
> Hi,
> 
> I made the experiment with httperf again, and used tcpdump to capture
> the packets on both server side and client side. I found that if there
> was some retransmission (e.t. client send several syn to server
> because it didn''t receive ack in time), the vif only receive the
last
> packet (syn) and missed the former ones. So I think there are some
> issues happened between the eth0(veth0) and vif in the dom0. But if I
> use a low latency scheduler written by myself (only modify the client
> scheduler and didn''t touch other parts ), there will not be
> retransmission or there is very few retransmission. I am not familiar
> with netback of xen. Can you give me some suggestion? Or which part
> source code I need to check to find the reason of packets loss between
> eth0(veth0) and vif in dom0? Thanks.
The netback code lives in either drivers/xen/netback or
drivers/net/xen-netback depending on the kernel you are using. The
interesting code is generally split between netback.c and interface.c.
>From the point of view of the kernel it is just a normal network driver.
For packets going from dom0->domU you can look at the .ndo_start_xmit
callback registered with the kernel (xenvif_start_xmit in mainline,
something else in some other kernels) and trace from there to see how
and where they can be dropped.

For packets coming the other way (domU->dom0) you can either start from
the place which calls netif_rx_ni() and work backwards or start from the
interrupt handler, i.e. the function passed to
bind_interdomain_evtchn_to_irqhandler() and work forwards.

You might also find starting from the *_tx_action and *_rx_action
functions useful starting points for tracing what is going on..

Ian.
> 
> Regards,
> Cong
> 
> 2011/9/30 Ian Campbell <Ian.Campbell@eu.citrix.com>:
> > On Fri, 2011-09-30 at 15:44 +0100, David Xu wrote:
> >> Hi,
> >>
> >> 2011/9/29 Ian Campbell <Ian.Campbell@citrix.com>:
> >> > On Fri, 2011-09-30 at 05:18 +0100, David Xu wrote:
> >> >> Hi,
> >> >>
> >> >> Does anybody know whether the ring buffer between front
end and back
> >> >> end will suffer from overflow? I just wonder if the ring
buffer will
> >> >> be full and drop some packets when the Net I/O load is
very heavy.
> >> >
> >> > In the case of networking whichever end is putting stuff on
the ring
> >> > checks that there is enough room and will stop the queue when
it cannot
> >> > transmit any more and restart when room becomes available.
> >>
> >> You mean even there is not enough room in ring buffer, xen will *
not
> >> drop the packets * and just delay the transmission.
> >
> > It''s not Xen but rather the kernel back and front ends which
is involved
> > here. You can examine the hard_start_xmit functions in both netback
and
> > netfront to determine for yourself whether or not packets can be
dropped
> > and when.
> >
> >>  I used httperf to
> >> measure the performance of web server running in a VM (The
workload in
> >> this VM is mixed, so it can not benefit from boost mechanism. The
net
> >> i/o suffers from relatively high latency which depends on the
number
> >> of VMs in the system). I found that with the increase of request
rate
> >> in client side, the connection rate will drop and the connection
time
> >> will increase dramatically. And the retransmission appears when
the
> >> request rate is over than a quantum. So I doubted that the
http/tcp
> >> connection suffer from the packets drop when the ring buffer is
full
> >> because of high request rate.
> >>
> >> >
> >> >> BTW, If I want to change the size of i/o ring buffer, how
should I do?
> >> >> I tried to reset the NET_TX_RING_SIZE and
NET_RX_RING_SIZE in both
> >> >> front end and back end, but it seems doesn''t
work. Thanks.
> >> >
> >> > Currently the rings are limited to 1 page so if you want to
increase the
> >> > size you would need to add multipage ring support to the
network
> >> > protocol. There have been patches to do this for the blk
protocol but I
> >> > do not recall any for the net protocol.
> >>
> >> Yes, increasing the size is relatively hard. So I just want reduce
the
> >> size of ring buffer to make sure my doubt described above. I
directly
> >> set  NET_TX_RING_SIZE and  NET_RX_RING_SIZE to 128, but it
doesn''t seem to work.
> >
> > You need to make sure both ends of the connection agree on the ring
> > size.
> >
> > I''m afraid this is not a very common thing to want to do so
if you want
> > to persist with this approach I''m afraid you''ll have
to do some
> > debugging.
> >
> > Ian.
> >
> >
> >


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hi,

I want to use the libxc function xc_sched_credit_domain_set( int
xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in a
user-level application. But I don''t know how to set the value of
xc_handle. Is there somebody familiar with it? Thanks.

Regards,
Cong

2011/10/4 David Xu <davidxu06@gmail.com>:
> Hi,
>
> I made the experiment with httperf again, and used tcpdump to capture
> the packets on both server side and client side. I found that if there
> was some retransmission (e.t. client send several syn to server
> because it didn''t receive ack in time), the vif only receive the
last
> packet (syn) and missed the former ones. So I think there are some
> issues happened between the eth0(veth0) and vif in the dom0. But if I
> use a low latency scheduler written by myself (only modify the client
> scheduler and didn''t touch other parts ), there will not be
> retransmission or there is very few retransmission. I am not familiar
> with netback of xen. Can you give me some suggestion? Or which part
> source code I need to check to find the reason of packets loss between
> eth0(veth0) and vif in dom0? Thanks.
>
> Regards,
> Cong
>
> 2011/9/30 Ian Campbell <Ian.Campbell@eu.citrix.com>:
>> On Fri, 2011-09-30 at 15:44 +0100, David Xu wrote:
>>> Hi,
>>>
>>> 2011/9/29 Ian Campbell <Ian.Campbell@citrix.com>:
>>> > On Fri, 2011-09-30 at 05:18 +0100, David Xu wrote:
>>> >> Hi,
>>> >>
>>> >> Does anybody know whether the ring buffer between front
end and back
>>> >> end will suffer from overflow? I just wonder if the ring
buffer will
>>> >> be full and drop some packets when the Net I/O load is
very heavy.
>>> >
>>> > In the case of networking whichever end is putting stuff on
the ring
>>> > checks that there is enough room and will stop the queue when
it cannot
>>> > transmit any more and restart when room becomes available.
>>>
>>> You mean even there is not enough room in ring buffer, xen will *
not
>>> drop the packets * and just delay the transmission.
>>
>> It''s not Xen but rather the kernel back and front ends which
is involved
>> here. You can examine the hard_start_xmit functions in both netback and
>> netfront to determine for yourself whether or not packets can be
dropped
>> and when.
>>
>>>  I used httperf to
>>> measure the performance of web server running in a VM (The workload
in
>>> this VM is mixed, so it can not benefit from boost mechanism. The
net
>>> i/o suffers from relatively high latency which depends on the
number
>>> of VMs in the system). I found that with the increase of request
rate
>>> in client side, the connection rate will drop and the connection
time
>>> will increase dramatically. And the retransmission appears when the
>>> request rate is over than a quantum. So I doubted that the http/tcp
>>> connection suffer from the packets drop when the ring buffer is
full
>>> because of high request rate.
>>>
>>> >
>>> >> BTW, If I want to change the size of i/o ring buffer, how
should I do?
>>> >> I tried to reset the NET_TX_RING_SIZE and NET_RX_RING_SIZE
in both
>>> >> front end and back end, but it seems doesn''t
work. Thanks.
>>> >
>>> > Currently the rings are limited to 1 page so if you want to
increase the
>>> > size you would need to add multipage ring support to the
network
>>> > protocol. There have been patches to do this for the blk
protocol but I
>>> > do not recall any for the net protocol.
>>>
>>> Yes, increasing the size is relatively hard. So I just want reduce
the
>>> size of ring buffer to make sure my doubt described above. I
directly
>>> set  NET_TX_RING_SIZE and  NET_RX_RING_SIZE to 128, but it
doesn''t seem to work.
>>
>> You need to make sure both ends of the connection agree on the ring
>> size.
>>
>> I''m afraid this is not a very common thing to want to do so if
you want
>> to persist with this approach I''m afraid you''ll have
to do some
>> debugging.
>>
>> Ian.
>>
>>
>>
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Stop top posting, I''ve asked you this before.

On Fri, 2011-10-07 at 16:23 +0100, David Xu wrote:
> Hi,
> 
> I want to use the libxc function xc_sched_credit_domain_set( int
> xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in a
> user-level application. But I don''t know how to set the value of
> xc_handle. Is there somebody familiar with it? Thanks.
You can''t use xc_* functions in the kernel, they are userspace. You
could in theory look at the implementation of xc_sched_credit_domain_set
to find out what that hypercall is and make the underlying hypercall
instead, But:

The hypercall in question here is a domctl which are not available /
allowed from kernel space (for policy and semantic reasons rather than
technological ones).

As I explained elsewhere I don''t think the architecture you have in
mind
is correct. I think you should look at exporting the necessary
statistics from the kernel to userspace and have a daemon implement the
logic and mechanisms for changing the scheduling parameters as required.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

At 16:49 +0100 on 07 Oct (1318006180), Ian Campbell
wrote:
> Stop top posting, I''ve asked you this before.
Also, please start a new thread for a new question; the subject line of
your email didn''t match the contents.
> On Fri, 2011-10-07 at 16:23 +0100, David Xu wrote:
> > Hi,
> > 
> > I want to use the libxc function xc_sched_credit_domain_set( int
> > xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in a
> > user-level application. But I don''t know how to set the value
of
> > xc_handle. Is there somebody familiar with it? Thanks.
> 
> You can''t use xc_* functions in the kernel, they are userspace.
He''s programming in userspace. :)

You need to call xc_interface_open() to get an xc_handle.  Please read
through the rest of xenctrl.h before asking any more coding questions
here.  You could also have found the answer by looking at any of the
other programs in the tools/ directore that use the xc interface. 

Cheers,

Tim.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hi,

This time I want to call it in the user-space. I wrote a user-level
application and call the xc_sched_credit_domain_set( int
 xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in
it. But I don''t know how to pass the xc_handle parameter in this
function. Thanks.

Regards,
Cong

2011/10/7 Ian Campbell
<Ian.Campbell@eu.citrix.com>:
> Stop top posting, I''ve asked you this before.
>
> On Fri, 2011-10-07 at 16:23 +0100, David Xu wrote:
>> Hi,
>>
>> I want to use the libxc function xc_sched_credit_domain_set( int
>> xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in a
>> user-level application. But I don''t know how to set the value
of
>> xc_handle. Is there somebody familiar with it? Thanks.
>
> You can''t use xc_* functions in the kernel, they are userspace.
You
> could in theory look at the implementation of xc_sched_credit_domain_set
> to find out what that hypercall is and make the underlying hypercall
> instead, But:
>
> The hypercall in question here is a domctl which are not available /
> allowed from kernel space (for policy and semantic reasons rather than
> technological ones).
>
> As I explained elsewhere I don''t think the architecture you have
in mind
> is correct. I think you should look at exporting the necessary
> statistics from the kernel to userspace and have a daemon implement the
> logic and mechanisms for changing the scheduling parameters as required.
>
> Ian.
>
>
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Fri, 2011-10-07 at 16:56 +0100, Tim Deegan wrote:
> At 16:49 +0100 on 07 Oct (1318006180), Ian Campbell wrote:
> > Stop top posting, I''ve asked you this before.
> 
> Also, please start a new thread for a new question; the subject line of
> your email didn''t match the contents.
> 
> > On Fri, 2011-10-07 at 16:23 +0100, David Xu wrote:
> > > Hi,
> > > 
> > > I want to use the libxc function xc_sched_credit_domain_set( int
> > > xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom)
in a
> > > user-level application. But I don''t know how to set the
value of
> > > xc_handle. Is there somebody familiar with it? Thanks.
> > 
> > You can''t use xc_* functions in the kernel, they are
userspace.
> 
> He''s programming in userspace. :)
So he is, I was confused by the existing context...

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Fri, 2011-10-07 at 17:23 +0100, David Xu wrote:
> Hi,
> 
> This time I want to call it in the user-space. I wrote a user-level
> application and call the xc_sched_credit_domain_set( int
>  xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in
> it. But I don''t know how to pass the xc_handle parameter in this
> function. Thanks.
Tim already answered this. But why do you persist in top posting?
You''ve
been asked not to _repeatedly_.

Please re-read http://wiki.xen.org/xenwiki/AskingXenDevelQuestions. In
particular the etiquette section.

Perhaps I should start ignoring top posted emails. :-/

Ian.
> 
> Regards,
> Cong
> 
> 2011/10/7 Ian Campbell <Ian.Campbell@eu.citrix.com>:
> > Stop top posting, I''ve asked you this before.
> >
> > On Fri, 2011-10-07 at 16:23 +0100, David Xu wrote:
> >> Hi,
> >>
> >> I want to use the libxc function xc_sched_credit_domain_set( int
> >> xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom)
in a
> >> user-level application. But I don''t know how to set the
value of
> >> xc_handle. Is there somebody familiar with it? Thanks.
> >
> > You can''t use xc_* functions in the kernel, they are
userspace. You
> > could in theory look at the implementation of
xc_sched_credit_domain_set
> > to find out what that hypercall is and make the underlying hypercall
> > instead, But:
> >
> > The hypercall in question here is a domctl which are not available /
> > allowed from kernel space (for policy and semantic reasons rather than
> > technological ones).
> >
> > As I explained elsewhere I don''t think the architecture you
have in mind
> > is correct. I think you should look at exporting the necessary
> > statistics from the kernel to userspace and have a daemon implement
the
> > logic and mechanisms for changing the scheduling parameters as
required.
> >
> > Ian.
> >
> >
> >


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Sorry, I did not notice his email and I did not mean to. I will be
more careful next time.

Regards,
Cong

2011/10/7 Ian Campbell <Ian.Campbell@citrix.com>:
> On Fri, 2011-10-07 at 17:23 +0100, David Xu wrote:
>> Hi,
>>
>> This time I want to call it in the user-space. I wrote a user-level
>> application and call the xc_sched_credit_domain_set( int
>>  xc_handle, uint32_t domid, struct xen_domctl_sched_credit *sdom) in
>> it. But I don''t know how to pass the xc_handle parameter in
this
>> function. Thanks.
>
> Tim already answered this. But why do you persist in top posting?
You''ve
> been asked not to _repeatedly_.
>
> Please re-read http://wiki.xen.org/xenwiki/AskingXenDevelQuestions. In
> particular the etiquette section.
>
> Perhaps I should start ignoring top posted emails. :-/
>
> Ian.
>
>>
>> Regards,
>> Cong
>>
>> 2011/10/7 Ian Campbell <Ian.Campbell@eu.citrix.com>:
>> > Stop top posting, I''ve asked you this before.
>> >
>> > On Fri, 2011-10-07 at 16:23 +0100, David Xu wrote:
>> >> Hi,
>> >>
>> >> I want to use the libxc function xc_sched_credit_domain_set(
int
>> >> xc_handle, uint32_t domid, struct xen_domctl_sched_credit
*sdom) in a
>> >> user-level application. But I don''t know how to set
the value of
>> >> xc_handle. Is there somebody familiar with it? Thanks.
>> >
>> > You can''t use xc_* functions in the kernel, they are
userspace. You
>> > could in theory look at the implementation of
xc_sched_credit_domain_set
>> > to find out what that hypercall is and make the underlying
hypercall
>> > instead, But:
>> >
>> > The hypercall in question here is a domctl which are not available
/
>> > allowed from kernel space (for policy and semantic reasons rather
than
>> > technological ones).
>> >
>> > As I explained elsewhere I don''t think the architecture
you have in mind
>> > is correct. I think you should look at exporting the necessary
>> > statistics from the kernel to userspace and have a daemon
implement the
>> > logic and mechanisms for changing the scheduling parameters as
required.
>> >
>> > Ian.
>> >
>> >
>> >
>
>
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel