[libvirt] [PATCH] Fix vm's outbound traffic control problem

[Eiichi Tsukata]

Hello,

This is a patch to fix vm's outbound traffic control problem.

Currently, vm's outbound traffic control by libvirt doesn't go well.
This problem was previously discussed at libvir-list ML, however
it seems that there isn't still any answer to the problem.
http://www.redhat.com/archives/libvir-list/2011-August/msg00333.html

I measured Guest(with virtio-net) to Host TCP throughput with the
command netperf -H.
Here are the outbound QoS parameters and the results.

outbound average rate[kilobytes/s] : Guest to Host throughput[Mbit/s]
==
1024  (8Mbit/s): 4.56
2048  (16Mbit/s)   : 3.29
4096  (32Mbit/s)   : 3.35
8192  (64Mbit/s)   : 3.95
16384 (128Mbit/s)  : 4.08
32768 (256Mbit/s)  : 3.94
65536 (512Mbit/s)  : 3.23

The outbound traffic goes down unreasonably and is even not controled.

The cause of this problem is too large mtu value in tc filter command run by
libvirt. The command uses burst value to set mtu and the burst is equal to
average rate value if it's not set. This value is too large. For example
if the average rate is set to 1024 kilobytes/s, the mtu value is set to 1024
kilobytes. That's too large compared to the size of network packets.
Here libvirt applies tc ingress filter to Host's vnet(tun) device.
Tc ingress filter is implemented with TBF(Token Buckets Filter) algorithm. TBF
uses mtu value to calculate the amount of token consumed by each packet. With 
too
large mtu value, the token consumption rate is set too large. This leads to
token starvation and deterioration of TCP throughput.

Then, should we use the default mtu value 2 kilobytes?
The anser is No, because Guest with virtio-net device uses 65536 bytes
as mtu to transmit packets to Host, and the tc filter with the default mtu
value 2k drops packets whose size is larger than 2k. So, the most packets
is droped and again leads to deterioration of TCP throughput.

The appropriate mtu value is 65536 bytes which is equal to the maximum value
of network interface device defined in linux/netdevice.h. The value is
not so large that it causes token starvation and not so small that it
drops most packets.
Therefore this patch set the mtu value to 64kb(== 65535 bytes).

Again, here are the outbound QoS parameters and the TCP throughput with
the libvirt patched.

outbound average rate[kilobytes/s] : Guest to Host throughput[Mbit/s]
==
1024  (8Mbit/s): 8.22
2048  (16Mbit/s)   : 16.42
4096  (32Mbit/s)   : 32.93
8192  (64Mbit/s)   : 66.85
16384 (128Mbit/s)  : 133.88
32768 (256Mbit/s)  : 271.01
65536 (512Mbit/s)  : 547.32

The outbound traffic conforms to the given limit.

Thank you,

Signed-off-by: Eiichi Tsukata eiichi.tsukata...@hitachi.com
---
 AUTHORS   |1 +
 src/util/virnetdevbandwidth.c |2 +-
 2 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/AUTHORS b/AUTHORS
index b876ee6..375db24 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -246,6 +246,7 @@ Patches have also been contributed by:
   Gerd Hoffmannkra...@redhat.com
   Viktor Mihajlovski   mihaj...@linux.vnet.ibm.com
   Thang Pham   thang.p...@us.ibm.com
+  Eiichi Tsukata   eiichi.tsukata...@hitachi.com

   [send patches to get your name here]

diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
index b9bd2e3..93df5c1 100644
--- a/src/util/virnetdevbandwidth.c
+++ b/src/util/virnetdevbandwidth.c
@@ -136,7 +136,7 @@ virNetDevBandwidthSet(const char *ifname,
 virCommandAddArgList(cmd, filter, add, dev, ifname, parent,
  :, protocol, ip, u32, match, ip,
  src, 0.0.0.0/0, police, rate, average,
- burst, burst, mtu, burst, drop, flowid,
+ burst, burst, mtu, 64kb, drop, flowid,
  :1, NULL);

 if (virCommandRun(cmd, NULL)  0)
-- 
1.7.4.4

-- 
Eiichi Tsukata
Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: eiichi.tsukata...@hitachi.com

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] build: define WITH_INTERFACE for the driver

[Osier Yang]

On 2012年06月29日 07:57, Eric Blake wrote:

Our code was mistakenly relying on an undefined macro, WITH_INTERFACE,
for determining whether to load the interface driver which wraps the
netcf library.  Clean this situation up by having only one automake
conditional for the driver, and having both WITH_NETCF (library
detected) and WITH_INTERFACE (driver enabled) in C code, in case a
future patch ever adds a network management via means other than
the netcf library.


Foresighted. :-)



While at it, output more information at the conclusion of configure
about the various drivers we enabled.

* configure.ac: Enhance with_netcf, and add with_interface.
Improve output to list final decisions.  Replace WITH_NETCF with
WITH_INTERFACE.
* src/interface/netcf_driver.c: Rename...
* src/interface/interface_driver.c: ...to this.
* src/interface/interface_driver.h: Likewise.
* daemon/Makefile.am (libvirtd_LDADD): Reflect better naming.
* src/Makefile.am (libvirt_driver_interface_la_*): Likewise.
(INTERFACE_DRIVER_SOURCES): Reflect file moves.
* daemon/libvirtd.c (daemonInitialize): Likewise.
* tools/virsh.c (vshShowVersion): Show both driver and library
decisions.
* libvirt.spec.in (with_interface): Tweak to deal with new usage
as a real switch.
---

I think this addresses the point that Osier raised here:
https://www.redhat.com/archives/libvir-list/2012-June/msg01266.html

but it is complex enough that I'd appreciate a careful review.

  configure.ac   |   44 
  daemon/Makefile.am |2 +-
  daemon/libvirtd.c  |6 +--
  libvirt.spec.in|   10 +++--
  src/Makefile.am|4 +-
  .../{netcf_driver.c =  interface_driver.c} |4 +-
  .../{netcf_driver.h =  interface_driver.h} |0
  tools/virsh.c  |   11 +++--
  8 files changed, 59 insertions(+), 22 deletions(-)
  rename src/interface/{netcf_driver.c =  interface_driver.c} (99%)
  rename src/interface/{netcf_driver.h =  interface_driver.h} (100%)

diff --git a/configure.ac b/configure.ac
index 6436885..a29b3b2 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1755,6 +1755,7 @@ if test $with_network = yes ; then
  fi
  AM_CONDITIONAL([WITH_NETWORK], [test $with_network = yes])

+dnl check whether helper code is needed for above selections
  with_bridge=no
  if test $with_qemu:$with_lxc:$with_network != no:no:no; then
  with_bridge=yes
@@ -1762,16 +1763,31 @@ if test $with_qemu:$with_lxc:$with_network != 
no:no:no; then
  fi
  AM_CONDITIONAL([WITH_BRIDGE], [test $with_bridge = yes])

-dnl netcf library
+dnl check if the interface driver should be compiled
+
+AC_ARG_WITH([interface],
+  AC_HELP_STRING([--with-interface],
+[with host interface driver @:@default=check@:@]),[],
+[with_interface=check])


Do we have to expose with-interface? It will give the user
a logic question, pick with-interface, or 'with-netcf', or
both, even more when we have other implementations of interface
driver in future. however, the logic is simple, and we do it
inside actually: as long as one implementation of the interface
driver is picked to compile, we have the WITH_INTERFACE. so IMHO
no need to give the user the simple logic question. :-)


+
+dnl there's no use compiling the interface driver without the libvirt daemon
+if test $with_libvirtd = no; then
+  with_interface=no
+fi


If we don't expose 'with-interface', we don't need this..


+
+dnl The interface driver depends on the netcf library
  AC_ARG_WITH([netcf],
AC_HELP_STRING([--with-netcf], [libnetcf support to configure physical host 
network interfaces @:@default=check@:@]),
  [], [with_netcf=check])

  NETCF_CFLAGS=
  NETCF_LIBS=
-if test $with_libvirtd = no ; then
+if test $with_libvirtd = no || test $with_interface = no; then
with_netcf=no
  fi
+if test $with_interface:$with_netcf = yes:check; then
+  with_netcf=yes
+fi
  if test $with_netcf = yes || test $with_netcf = check; then
PKG_CHECK_MODULES(NETCF, netcf= $NETCF_REQUIRED,
  [with_netcf=yes], [
@@ -1792,11 +1808,21 @@ if test $with_netcf = yes || test $with_netcf = 
check; then
  fi
fi
  fi
-AM_CONDITIONAL([WITH_NETCF], [test $with_netcf = yes])
  AC_SUBST([NETCF_CFLAGS])
  AC_SUBST([NETCF_LIBS])

+dnl Final decision on the interface driver
+if test $with_interface = check; then
+  with_interface=$with_netcf
+fi
+
+if test $with_interface = yes ; then
+  AC_DEFINE_UNQUOTED([WITH_INTERFACE], [1],
+[whether interface driver is enabled])
+fi
+AM_CONDITIONAL([WITH_INTERFACE], [test $with_interface = yes])



And above changes, what we need is just:

if test $with_netfs = yes || with_something_else = yes; then
AC_DEFINE_UNQUOTED([WITH_INTERFACE], [1],
  [whether interface driver is enabled])
fi
AM_CONDITIONAL([WITH_INTERFACE], [test $with_interface = yes])


+dnl Check whether the 

Re: [libvirt] [PATCH] Fix vm's outbound traffic control problem

[Michal Privoznik]

On 29.06.2012 08:09, Eiichi Tsukata wrote:
 Hello,
 
 This is a patch to fix vm's outbound traffic control problem.
 
 Currently, vm's outbound traffic control by libvirt doesn't go well.
 This problem was previously discussed at libvir-list ML, however
 it seems that there isn't still any answer to the problem.
 http://www.redhat.com/archives/libvir-list/2011-August/msg00333.html
 
 I measured Guest(with virtio-net) to Host TCP throughput with the
 command netperf -H.
 Here are the outbound QoS parameters and the results.
 
 outbound average rate[kilobytes/s] : Guest to Host throughput[Mbit/s]
 ==
 1024  (8Mbit/s): 4.56
 2048  (16Mbit/s)   : 3.29
 4096  (32Mbit/s)   : 3.35
 8192  (64Mbit/s)   : 3.95
 16384 (128Mbit/s)  : 4.08
 32768 (256Mbit/s)  : 3.94
 65536 (512Mbit/s)  : 3.23
 
 The outbound traffic goes down unreasonably and is even not controled.
 
 The cause of this problem is too large mtu value in tc filter command run by
 libvirt. The command uses burst value to set mtu and the burst is equal to
 average rate value if it's not set. This value is too large. For example
 if the average rate is set to 1024 kilobytes/s, the mtu value is set to 1024
 kilobytes. That's too large compared to the size of network packets.
 Here libvirt applies tc ingress filter to Host's vnet(tun) device.
 Tc ingress filter is implemented with TBF(Token Buckets Filter) algorithm. TBF
 uses mtu value to calculate the amount of token consumed by each packet. With 
 too
 large mtu value, the token consumption rate is set too large. This leads to
 token starvation and deterioration of TCP throughput.
 
 Then, should we use the default mtu value 2 kilobytes?
 The anser is No, because Guest with virtio-net device uses 65536 bytes
 as mtu to transmit packets to Host, and the tc filter with the default mtu
 value 2k drops packets whose size is larger than 2k. So, the most packets
 is droped and again leads to deterioration of TCP throughput.
 
 The appropriate mtu value is 65536 bytes which is equal to the maximum value
 of network interface device defined in linux/netdevice.h. The value is
 not so large that it causes token starvation and not so small that it
 drops most packets.
 Therefore this patch set the mtu value to 64kb(== 65535 bytes).
 
 Again, here are the outbound QoS parameters and the TCP throughput with
 the libvirt patched.
 
 outbound average rate[kilobytes/s] : Guest to Host throughput[Mbit/s]
 ==
 1024  (8Mbit/s): 8.22
 2048  (16Mbit/s)   : 16.42
 4096  (32Mbit/s)   : 32.93
 8192  (64Mbit/s)   : 66.85
 16384 (128Mbit/s)  : 133.88
 32768 (256Mbit/s)  : 271.01
 65536 (512Mbit/s)  : 547.32
 
 The outbound traffic conforms to the given limit.
 
 Thank you,
 
 Signed-off-by: Eiichi Tsukata eiichi.tsukata...@hitachi.com
 ---
  AUTHORS   |1 +
  src/util/virnetdevbandwidth.c |2 +-
  2 files changed, 2 insertions(+), 1 deletions(-)
 
 diff --git a/AUTHORS b/AUTHORS
 index b876ee6..375db24 100644
 --- a/AUTHORS
 +++ b/AUTHORS
 @@ -246,6 +246,7 @@ Patches have also been contributed by:
Gerd Hoffmannkra...@redhat.com
Viktor Mihajlovski   mihaj...@linux.vnet.ibm.com
Thang Pham   thang.p...@us.ibm.com
 +  Eiichi Tsukata   eiichi.tsukata...@hitachi.com
 
[send patches to get your name here]
 
 diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
 index b9bd2e3..93df5c1 100644
 --- a/src/util/virnetdevbandwidth.c
 +++ b/src/util/virnetdevbandwidth.c
 @@ -136,7 +136,7 @@ virNetDevBandwidthSet(const char *ifname,
  virCommandAddArgList(cmd, filter, add, dev, ifname, parent,
   :, protocol, ip, u32, match, ip,
   src, 0.0.0.0/0, police, rate, average,
 - burst, burst, mtu, burst, drop, flowid,
 + burst, burst, mtu, 64kb, drop, flowid,
   :1, NULL);
 
  if (virCommandRun(cmd, NULL)  0)
 

Yeah, this tc magic was the hardest part when I've introduced QoS
feature. I tested it on small values where this has not fully showed.

ACK and pushed. Thanks!

Michal

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] Entering freeze for libvirt-0.9.13

[Guido Günther]

On Mon, Jun 25, 2012 at 07:20:59PM +0800, Daniel Veillard wrote:
   I just tagged the git tree and I pushed the tarball for rc1:
 ftp://libvirt.org/libvirt/libvirt-0.9.13-rc1.tar.gz
 
 the rpms I build are coming along soon.
   Please give it a try, the basic seems okay to me (kvm driver seems to
 need the qemu one now but otherwise, no surprize so far)
 Give it a try too, and let's find if there are portability issues too !
 I think i would make an rc2 release on wed or thursday, and then if
 everything is fine the release next week-end or on Monday.

Build looks good so far on the Debian Buildds:


https://buildd.debian.org/status/package.php?p=libvirtsuite=experimental

Cheers,
 -- Guido

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] Intend to add fuse filesystem support for libvirt lxc

[Daniel P. Berrange]

On Fri, Jun 29, 2012 at 11:05:33AM +0800, Gao feng wrote:
 Hi Everyone,
 
 Now I'm working on making the container's system info(such as 
 /proc/meminfo,cpuinfo..) isolate from
 the host.
 
 I made a patch which implement showing the /proc/meminfo base on container's 
 memcg,
 and sent it to the 
 community.(http://marc.info/?l=linux-mmm=133826035821338w=2)
 
 but I found it's difficult to be accepted, because this way is ugly,
 and somebody gave me some suggestions.
 
 the first way is making another kernel file(just like 
 memory.limit_in_bytes),and mount it to the container.
 I don't like this way,because there will be many redundance information 
 between this new kernel file and
 the existing kernel file,some files such as memory.stat already contains 
 memcg information.
 
 the other way is adding fuse filesystem support for libvirt lxc.
 with this way,
 we can simply collect information from cgroup in fuse_operations.read 
 function,
 and mount this file to the container.
 we can impletment isolate meminfo in userspace without changing kernel codes.
 
 I have impletment fuse support for libvrit now, and ready to impletement the 
 meminfo isolated.
 
 I want to know if you have any comment or another ideas?

Having read that long thread you quote above, I tend to agree that
creating a FUSE filesystem is probably the best option we have for
dealing with /proc in the short term.  I'd expect it to be provided
either in the libvirt_lxc  controller process that is forked per
container, or even perhaps in a new process libvirt_lxc_fuse forked
per container.

I can see two ways todo it

  - Provide a FUSE filesystem that just contains the couple of files
we need, and then bind them over the top of the files in /proc

  - Provide a FUSE filesystem that does a union-fs like layering on
top of an existing /proc mount, avoiding the need for bind mounts


Daniel
-- 
|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org  -o- http://virt-manager.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] Fix vm's outbound traffic control problem

[Daniel Veillard]

On Fri, Jun 29, 2012 at 11:08:18AM +0200, Michal Privoznik wrote:
 On 29.06.2012 08:09, Eiichi Tsukata wrote:
  Hello,
  
  This is a patch to fix vm's outbound traffic control problem.
  
  Currently, vm's outbound traffic control by libvirt doesn't go well.
  This problem was previously discussed at libvir-list ML, however
  it seems that there isn't still any answer to the problem.
  http://www.redhat.com/archives/libvir-list/2011-August/msg00333.html
  
  I measured Guest(with virtio-net) to Host TCP throughput with the
  command netperf -H.
  Here are the outbound QoS parameters and the results.
  
  outbound average rate[kilobytes/s] : Guest to Host throughput[Mbit/s]
  ==
  1024  (8Mbit/s): 4.56
  2048  (16Mbit/s)   : 3.29
  4096  (32Mbit/s)   : 3.35
  8192  (64Mbit/s)   : 3.95
  16384 (128Mbit/s)  : 4.08
  32768 (256Mbit/s)  : 3.94
  65536 (512Mbit/s)  : 3.23
  
  The outbound traffic goes down unreasonably and is even not controled.
  
  The cause of this problem is too large mtu value in tc filter command run 
  by
  libvirt. The command uses burst value to set mtu and the burst is equal to
  average rate value if it's not set. This value is too large. For example
  if the average rate is set to 1024 kilobytes/s, the mtu value is set to 1024
  kilobytes. That's too large compared to the size of network packets.
  Here libvirt applies tc ingress filter to Host's vnet(tun) device.
  Tc ingress filter is implemented with TBF(Token Buckets Filter) algorithm. 
  TBF
  uses mtu value to calculate the amount of token consumed by each packet. 
  With too
  large mtu value, the token consumption rate is set too large. This leads to
  token starvation and deterioration of TCP throughput.
  
  Then, should we use the default mtu value 2 kilobytes?
  The anser is No, because Guest with virtio-net device uses 65536 bytes
  as mtu to transmit packets to Host, and the tc filter with the default mtu
  value 2k drops packets whose size is larger than 2k. So, the most packets
  is droped and again leads to deterioration of TCP throughput.
  
  The appropriate mtu value is 65536 bytes which is equal to the maximum value
  of network interface device defined in linux/netdevice.h. The value is
  not so large that it causes token starvation and not so small that it
  drops most packets.
  Therefore this patch set the mtu value to 64kb(== 65535 bytes).
  
  Again, here are the outbound QoS parameters and the TCP throughput with
  the libvirt patched.
  
  outbound average rate[kilobytes/s] : Guest to Host throughput[Mbit/s]
  ==
  1024  (8Mbit/s): 8.22
  2048  (16Mbit/s)   : 16.42
  4096  (32Mbit/s)   : 32.93
  8192  (64Mbit/s)   : 66.85
  16384 (128Mbit/s)  : 133.88
  32768 (256Mbit/s)  : 271.01
  65536 (512Mbit/s)  : 547.32
  
  The outbound traffic conforms to the given limit.
  
  Thank you,
  
  Signed-off-by: Eiichi Tsukata eiichi.tsukata...@hitachi.com
  ---
   AUTHORS   |1 +
   src/util/virnetdevbandwidth.c |2 +-
   2 files changed, 2 insertions(+), 1 deletions(-)
  
  diff --git a/AUTHORS b/AUTHORS
  index b876ee6..375db24 100644
  --- a/AUTHORS
  +++ b/AUTHORS
  @@ -246,6 +246,7 @@ Patches have also been contributed by:
 Gerd Hoffmannkra...@redhat.com
 Viktor Mihajlovski   mihaj...@linux.vnet.ibm.com
 Thang Pham   thang.p...@us.ibm.com
  +  Eiichi Tsukata   eiichi.tsukata...@hitachi.com
  
 [send patches to get your name here]
  
  diff --git a/src/util/virnetdevbandwidth.c b/src/util/virnetdevbandwidth.c
  index b9bd2e3..93df5c1 100644
  --- a/src/util/virnetdevbandwidth.c
  +++ b/src/util/virnetdevbandwidth.c
  @@ -136,7 +136,7 @@ virNetDevBandwidthSet(const char *ifname,
   virCommandAddArgList(cmd, filter, add, dev, ifname, parent,
:, protocol, ip, u32, match, 
  ip,
src, 0.0.0.0/0, police, rate, average,
  - burst, burst, mtu, burst, drop, 
  flowid,
  + burst, burst, mtu, 64kb, drop, 
  flowid,
:1, NULL);
  
   if (virCommandRun(cmd, NULL)  0)
  
 
 Yeah, this tc magic was the hardest part when I've introduced QoS
 feature. I tested it on small values where this has not fully showed.
 
 ACK and pushed. Thanks!

  And special thanks for a very detailed explanation :-), excellent !

Daniel

-- 
Daniel Veillard  | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
dan...@veillard.com  | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library  http://libvirt.org/

--

Re: [libvirt] Entering freeze for libvirt-0.9.13

[Daniel Veillard]

On Fri, Jun 29, 2012 at 11:15:02AM +0200, Guido Günther wrote:
 On Mon, Jun 25, 2012 at 07:20:59PM +0800, Daniel Veillard wrote:
I just tagged the git tree and I pushed the tarball for rc1:
  ftp://libvirt.org/libvirt/libvirt-0.9.13-rc1.tar.gz
  
  the rpms I build are coming along soon.
Please give it a try, the basic seems okay to me (kvm driver seems to
  need the qemu one now but otherwise, no surprize so far)
  Give it a try too, and let's find if there are portability issues too !
  I think i would make an rc2 release on wed or thursday, and then if
  everything is fine the release next week-end or on Monday.
 
 Build looks good so far on the Debian Buildds:
 
   
 https://buildd.debian.org/status/package.php?p=libvirtsuite=experimental

  Thanks Guido !

at least the s390 patches didn't break code portability :-)

Daniel

-- 
Daniel Veillard  | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
dan...@veillard.com  | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library  http://libvirt.org/

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCH 0/6 v3] Support forward mode='hostdev' and interface pools

[Shradha Shah]

This patch series supports the forward mode='hostdev'. The functionality
of this mode is the same as interface type='hostdev' but with the added
benefit of using interface pools.

The patch series also contains a patch to support use of interface names
and PCI device addresses interchangeably in a network xml, and return
the appropriate one in actualDevice when networkAllocateActualDevice is
called.

At the top level managed attribute can be specified with identical results 
as when it's specified for a hostdev.

Currently forward mode='hostdev' does not support USB devices.

Shradha Shah (6):
  Prerequisite Patch. virDomainDevicePCIAddress and respective
functions moved to a new file called conf/device_conf.ch
  Moved the code to create implicit interface pool from PF to a new
function
  RNG updates, new xml parser/formatter code to support forward
mode=hostdev
  Code to return interface name or pci_addr of the VF in actualDevice
  Forward Mode Hostdev network driver Implementation
  Forward Mode 'Hostdev' qemu driver implementation

 docs/formatnetwork.html.in |   62 ++
 docs/schemas/network.rng   |   82 -
 include/libvirt/virterror.h|1 +
 src/Makefile.am|7 +-
 src/conf/device_conf.c |  135 +
 src/conf/device_conf.h |   65 +++
 src/conf/domain_conf.c |  114 ++--
 src/conf/domain_conf.h |   25 +---
 src/conf/network_conf.c|  126 +++--
 src/conf/network_conf.h|   29 +++-
 src/libvirt_private.syms   |   10 +-
 src/network/bridge_driver.c|  322 +---
 src/qemu/qemu_command.c|   27 ++-
 src/qemu/qemu_hotplug.c|7 +-
 src/qemu/qemu_monitor.c|   14 +-
 src/qemu/qemu_monitor.h|   17 +-
 src/qemu/qemu_monitor_json.c   |   14 +-
 src/qemu/qemu_monitor_json.h   |   14 +-
 src/qemu/qemu_monitor_text.c   |   16 +-
 src/qemu/qemu_monitor_text.h   |   14 +-
 src/util/virnetdev.c   |   29 ++--
 src/util/virnetdev.h   |4 +-
 src/xen/xend_internal.c|3 +-
 tests/networkxml2xmlin/hostdev-pf.xml  |   11 +
 tests/networkxml2xmlin/hostdev.xml |   10 +
 tests/networkxml2xmlout/hostdev-pf.xml |7 +
 tests/networkxml2xmlout/hostdev.xml|   10 +
 tests/networkxml2xmltest.c |2 +
 28 files changed, 890 insertions(+), 287 deletions(-)
 create mode 100644 src/conf/device_conf.c
 create mode 100644 src/conf/device_conf.h
 create mode 100644 tests/networkxml2xmlin/hostdev-pf.xml
 create mode 100644 tests/networkxml2xmlin/hostdev.xml
 create mode 100644 tests/networkxml2xmlout/hostdev-pf.xml
 create mode 100644 tests/networkxml2xmlout/hostdev.xml

-- 
1.7.4.4

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCH 1/6 v3] Prerequisite Patch. virDomainDevicePCIAddress and respective functions moved to a new file called conf/device_conf.ch

[Shradha Shah]

Refactoring existing code without causing any functional changes to prepare for 
new code.
This patch makes the code reusable.

Signed-off-by: Shradha Shah ss...@solarflare.com
---
 include/libvirt/virterror.h  |1 +
 src/Makefile.am  |7 ++-
 src/conf/device_conf.c   |  135 ++
 src/conf/device_conf.h   |   65 
 src/conf/domain_conf.c   |  114 ---
 src/conf/domain_conf.h   |   25 +---
 src/libvirt_private.syms |   10 ++-
 src/qemu/qemu_command.c  |   13 ++--
 src/qemu/qemu_hotplug.c  |7 +-
 src/qemu/qemu_monitor.c  |   14 ++--
 src/qemu/qemu_monitor.h  |   17 +++---
 src/qemu/qemu_monitor_json.c |   14 ++--
 src/qemu/qemu_monitor_json.h |   14 ++--
 src/qemu/qemu_monitor_text.c |   16 +++---
 src/qemu/qemu_monitor_text.h |   14 ++--
 src/xen/xend_internal.c  |3 +-
 16 files changed, 289 insertions(+), 180 deletions(-)

diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h
index 0e0bc9c..7ad1201 100644
--- a/include/libvirt/virterror.h
+++ b/include/libvirt/virterror.h
@@ -97,6 +97,7 @@ typedef enum {
 VIR_FROM_URI = 45,  /* Error from URI handling */
 VIR_FROM_AUTH = 46, /* Error from auth handling */
 VIR_FROM_DBUS = 47, /* Error from DBus */
+VIR_FROM_DEVICE = 48,   /* Error from Device */
 
 # ifdef VIR_ENUM_SENTINELS
 VIR_ERR_DOMAIN_LAST
diff --git a/src/Makefile.am b/src/Makefile.am
index 2309984..7ffb3c2 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -199,6 +199,9 @@ CONSOLE_CONF_SOURCES =  
\
 DOMAIN_LIST_SOURCES =  \
conf/virdomainlist.c conf/virdomainlist.h
 
+DEVICE_CONF_SOURCES =   \
+   conf/device_conf.c conf/device_conf.h
+
 CONF_SOURCES = \
$(NETDEV_CONF_SOURCES)  \
$(DOMAIN_CONF_SOURCES)  \
@@ -212,7 +215,8 @@ CONF_SOURCES =  
\
$(SECRET_CONF_SOURCES)  \
$(CPU_CONF_SOURCES) \
$(CONSOLE_CONF_SOURCES) \
-   $(DOMAIN_LIST_SOURCES)
+   $(DOMAIN_LIST_SOURCES)  \
+   $(DEVICE_CONF_SOURCES)
 
 # The remote RPC driver, covering domains, storage, networks, etc
 REMOTE_DRIVER_GENERATED = \
@@ -1526,6 +1530,7 @@ libvirt_lxc_SOURCES = 
\
$(ENCRYPTION_CONF_SOURCES)  \
$(NETDEV_CONF_SOURCES)  \
$(DOMAIN_CONF_SOURCES)  \
+   $(DEVICE_CONF_SOURCES)  \
$(SECRET_CONF_SOURCES)  \
$(CPU_CONF_SOURCES) \
$(SECURITY_DRIVER_SOURCES)  \
diff --git a/src/conf/device_conf.c b/src/conf/device_conf.c
new file mode 100644
index 000..d4eb764
--- /dev/null
+++ b/src/conf/device_conf.c
@@ -0,0 +1,135 @@
+/*
+ * device_conf.h: device XML handling
+ *
+ * Copyright (C) 2006-2012 Red Hat, Inc.
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307  USA
+ *
+ * Author: Shradha Shah ss...@solarflare.com
+ */
+
+#include config.h
+#include virterror_internal.h
+#include datatypes.h
+#include memory.h
+#include xml.h
+#include uuid.h
+#include util.h
+#include buf.h
+#include conf/device_conf.h
+
+#define VIR_FROM_THIS VIR_FROM_DEVICE
+
+#define virDeviceReportError(code, ...)  \
+virReportErrorHelper(VIR_FROM_DEVICE, code, __FILE__,\
+ __FUNCTION__, __LINE__, __VA_ARGS__)
+
+VIR_ENUM_IMPL(virDeviceAddressPciMulti,
+  VIR_DEVICE_ADDRESS_PCI_MULTI_LAST,
+  default,
+  on,
+  off)
+
+int virDevicePCIAddressIsValid(virDevicePCIAddressPtr addr)
+{
+/* PCI bus has 32 slots and 8 functions 

[libvirt] [PATCH 2/6 v3] Moved the code to create implicit interface pool from PF to a new function

[Shradha Shah]

Just code movement no functional changes here. This makes the code reusable

Signed-off-by: Shradha Shah ss...@solarflare.com
---
 src/network/bridge_driver.c |   84 ++
 1 files changed, 52 insertions(+), 32 deletions(-)

diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index 7e8de19..36afa1b 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -2730,6 +2730,56 @@ int networkRegister(void) {
  * backend function table.
  */
 
+/* networkCreateInterfacePool:
+ * @netdef: the original NetDef from the network
+ *
+ * Creates an implicit interface pool of VF's when a PF dev is given
+ */
+static int 
+networkCreateInterfacePool(virNetworkDefPtr netdef) {
+unsigned int num_virt_fns = 0;
+char **vfname = NULL;
+int ret = -1, ii = 0;
+
+if ((virNetDevGetVirtualFunctions(netdef-forwardPfs-dev,
+  vfname, num_virt_fns))  0) {
+networkReportError(VIR_ERR_INTERNAL_ERROR,
+   _(Could not get Virtual functions on %s),
+   netdef-forwardPfs-dev);
+goto finish;
+}
+
+if (num_virt_fns == 0) {
+networkReportError(VIR_ERR_INTERNAL_ERROR,
+   _(No Vf's present on SRIOV PF %s),
+   netdef-forwardPfs-dev);
+   goto finish;
+}
+
+if ((VIR_ALLOC_N(netdef-forwardIfs, num_virt_fns))  0) {
+virReportOOMError();
+goto finish;
+}
+
+netdef-nForwardIfs = num_virt_fns;
+
+for (ii = 0; ii  netdef-nForwardIfs; ii++) {
+netdef-forwardIfs[ii].dev = strdup(vfname[ii]);
+if (!netdef-forwardIfs[ii].dev) {
+virReportOOMError();
+goto finish;
+}
+netdef-forwardIfs[ii].usageCount = 0;
+}
+
+ret = 0;
+finish:
+for (ii = 0; ii  num_virt_fns; ii++)
+VIR_FREE(vfname[ii]);
+VIR_FREE(vfname);
+return ret;
+}
+
 /* networkAllocateActualDevice:
  * @iface: the original NetDef from the domain
  *
@@ -2748,8 +2798,6 @@ networkAllocateActualDevice(virDomainNetDefPtr iface)
 virNetworkObjPtr network;
 virNetworkDefPtr netdef;
 virPortGroupDefPtr portgroup;
-unsigned int num_virt_fns = 0;
-char **vfname = NULL;
 int ii;
 int ret = -1;
 
@@ -2895,36 +2943,11 @@ networkAllocateActualDevice(virDomainNetDefPtr iface)
  */
 if (netdef-forwardType == VIR_NETWORK_FORWARD_PASSTHROUGH) {
 if ((netdef-nForwardPfs  0)  (netdef-nForwardIfs = 0)) {
-if ((virNetDevGetVirtualFunctions(netdef-forwardPfs-dev,
-  vfname, num_virt_fns)) 
 0) {
+if ((networkCreateInterfacePool(netdef))  0) {
 networkReportError(VIR_ERR_INTERNAL_ERROR,
-   _(Could not get Virtual functions 
on %s),
-   netdef-forwardPfs-dev);
+   _(Could not Interface Pool));
 goto cleanup;
 }
-
-if (num_virt_fns == 0) {
-networkReportError(VIR_ERR_INTERNAL_ERROR,
-   _(No Vf's present on SRIOV PF %s),
-   netdef-forwardPfs-dev);
-goto cleanup;
-}
-
-if ((VIR_ALLOC_N(netdef-forwardIfs, num_virt_fns))  0) {
-virReportOOMError();
-goto cleanup;
-}
-
-netdef-nForwardIfs = num_virt_fns;
-
-for (ii = 0; ii  netdef-nForwardIfs; ii++) {
-netdef-forwardIfs[ii].dev = strdup(vfname[ii]);
-if (!netdef-forwardIfs[ii].dev) {
-virReportOOMError();
-goto cleanup;
-}
-netdef-forwardIfs[ii].usageCount = 0;
-}
 }
 
 /* pick first dev with 0 usageCount */
@@ -2976,9 +2999,6 @@ networkAllocateActualDevice(virDomainNetDefPtr iface)
 
 ret = 0;
 cleanup:
-for (ii = 0; ii  num_virt_fns; ii++)
-VIR_FREE(vfname[ii]);
-VIR_FREE(vfname);
 if (network)
 virNetworkObjUnlock(network);
 if (ret  0) {
-- 
1.7.4.4


--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCH 3/6 v3] RNG updates, new xml parser/formatter code to support forward mode=hostdev

[Shradha Shah]

This patch introduces the new forward mode='hostdev' along with attribute 
managed
Includes updates to the network RNG and new xml parser/formatter code.

Signed-off-by: Shradha Shah ss...@solarflare.com
---
 docs/schemas/network.rng   |   82 +++--
 src/conf/network_conf.c|  126 
 src/conf/network_conf.h|   29 +++-
 src/network/bridge_driver.c|   18 ++--
 tests/networkxml2xmlin/hostdev-pf.xml  |   11 +++
 tests/networkxml2xmlin/hostdev.xml |   10 +++
 tests/networkxml2xmlout/hostdev-pf.xml |7 ++
 tests/networkxml2xmlout/hostdev.xml|   10 +++
 tests/networkxml2xmltest.c |2 +
 9 files changed, 262 insertions(+), 33 deletions(-)

diff --git a/docs/schemas/network.rng b/docs/schemas/network.rng
index 2ae879e..d1297cd 100644
--- a/docs/schemas/network.rng
+++ b/docs/schemas/network.rng
@@ -82,17 +82,41 @@
   valuepassthrough/value
   valueprivate/value
   valuevepa/value
+  valuehostdev/value
+/choice
+  /attribute
+/optional
+
+optional
+  attribute name=managed
+choice
+  valueyes/value
+  valueno/value
 /choice
   /attribute
 /optional
 interleave
-  zeroOrMore
-element name='interface'
-  attribute name='dev'
-ref name='deviceName'/
-  /attribute
-/element
-  /zeroOrMore
+  choice
+group
+  zeroOrMore
+element name='interface'
+  attribute name='dev'
+ref name='deviceName'/
+  /attribute
+/element
+  /zeroOrMore
+/group
+group
+  zeroOrMore
+element name='address'
+  attribute name='type'
+valuepci/value
+  /attribute
+  ref name=pciaddress/
+/element
+  /zeroOrMore
+/group
+  /choice
   optional
 element name='pf'
   attribute name='dev'
@@ -238,4 +262,48 @@
   /interleave
 /element
   /define
+  define name=pciaddress
+optional
+  attribute name=domain
+ref name=pciDomain/
+  /attribute
+/optional
+attribute name=bus
+  ref name=pciBus/
+/attribute
+attribute name=slot
+  ref name=pciSlot/
+/attribute
+attribute name=function
+  ref name=pciFunc/
+/attribute
+optional
+  attribute name=multifunction
+choice
+  valueon/value
+  valueoff/value
+/choice
+  /attribute
+/optional
+  /define
+  define name=pciDomain
+data type=string
+  param name=pattern(0x)?[0-9a-fA-F]{1,4}/param
+/data
+  /define
+  define name=pciBus
+data type=string
+  param name=pattern(0x)?[0-9a-fA-F]{1,2}/param
+/data
+  /define
+  define name=pciSlot
+data type=string
+  param name=pattern(0x)?[0-1]?[0-9a-fA-F]/param
+/data
+  /define
+  define name=pciFunc
+data type=string
+  param name=pattern(0x)?[0-7]/param
+/data
+  /define
 /grammar
diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c
index 515bc36..be37856 100644
--- a/src/conf/network_conf.c
+++ b/src/conf/network_conf.c
@@ -48,10 +48,14 @@
 #define VIR_FROM_THIS VIR_FROM_NETWORK
 
 VIR_ENUM_DECL(virNetworkForward)
-
 VIR_ENUM_IMPL(virNetworkForward,
   VIR_NETWORK_FORWARD_LAST,
-  none, nat, route, bridge, private, vepa, 
passthrough )
+  none, nat, route, bridge, private, vepa, 
passthrough, hostdev)
+
+VIR_ENUM_DECL(virNetworkForwardHostdevDevice)
+VIR_ENUM_IMPL(virNetworkForwardHostdevDevice,
+  VIR_NETWORK_FORWARD_HOSTDEV_DEVICE_LAST,
+  none, pci)
 
 #define virNetworkReportError(code, ...)\
 virReportErrorHelper(VIR_FROM_NETWORK, code, __FILE__,  \
@@ -100,6 +104,12 @@ virPortGroupDefClear(virPortGroupDefPtr def)
 static void
 virNetworkForwardIfDefClear(virNetworkForwardIfDefPtr def)
 {
+VIR_FREE(def-device.dev);
+}
+
+static void
+virNetworkForwardPfDefClear(virNetworkForwardPfDefPtr def)
+{
 VIR_FREE(def-dev);
 }
 
@@ -163,12 +173,13 @@ void virNetworkDefFree(virNetworkDefPtr def)
 VIR_FREE(def-domain);
 
 for (ii = 0 ; ii  def-nForwardPfs  def-forwardPfs ; ii++) {
-virNetworkForwardIfDefClear(def-forwardPfs[ii]);
+virNetworkForwardPfDefClear(def-forwardPfs[ii]);
 }
 VIR_FREE(def-forwardPfs);
 
 for (ii = 0 ; ii  def-nForwardIfs  def-forwardIfs ; ii++) {
-

[libvirt] [PATCH 4/6 v3] Code to return interface name or pci_addr of the VF in actualDevice

[Shradha Shah]

The network pool should be able to keep track of both, network device
names nad PCI addresses, and return the appropriate one in the actualDevice
when networkAllocateActualDevice is called.

Signed-off-by: Shradha Shah ss...@solarflare.com
---
 src/network/bridge_driver.c |   33 +++--
 src/util/virnetdev.c|   29 -
 src/util/virnetdev.h|4 +++-
 3 files changed, 46 insertions(+), 20 deletions(-)

diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index 230012c..2f8a937 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -59,6 +59,7 @@
 #include dnsmasq.h
 #include configmake.h
 #include virnetdev.h
+#include pci.h
 #include virnetdevbridge.h
 #include virnetdevtap.h
 
@@ -2739,10 +2740,11 @@ static int
 networkCreateInterfacePool(virNetworkDefPtr netdef) {
 unsigned int num_virt_fns = 0;
 char **vfname = NULL;
+struct pci_config_address **virt_fns;
 int ret = -1, ii = 0;
 
 if ((virNetDevGetVirtualFunctions(netdef-forwardPfs-dev,
-  vfname, num_virt_fns))  0) {
+  vfname, virt_fns, num_virt_fns))  0) 
{
 networkReportError(VIR_ERR_INTERNAL_ERROR,
_(Could not get Virtual functions on %s),
netdef-forwardPfs-dev);
@@ -2764,19 +2766,38 @@ networkCreateInterfacePool(virNetworkDefPtr netdef) {
 netdef-nForwardIfs = num_virt_fns;
 
 for (ii = 0; ii  netdef-nForwardIfs; ii++) {
-netdef-forwardIfs[ii].device.dev = strdup(vfname[ii]);
-if (!netdef-forwardIfs[ii].device.dev) {
-virReportOOMError();
-goto finish;
+if (netdef-forwardType == VIR_NETWORK_FORWARD_PASSTHROUGH) {
+if(vfname[ii]) {
+netdef-forwardIfs[ii].device.dev = strdup(vfname[ii]);
+if (!netdef-forwardIfs[ii].device.dev) {
+virReportOOMError();
+goto finish;
+}
+}
+else {
+networkReportError(VIR_ERR_INTERNAL_ERROR,
+   _(Passthrough mode requires interface 
names));
+goto finish;
+}
+}
+else if (netdef-forwardType == VIR_NETWORK_FORWARD_HOSTDEV) {
+netdef-forwardIfs[ii].type = 
VIR_NETWORK_FORWARD_HOSTDEV_DEVICE_PCI; /*Assuming PCI as VF's are PCI devices 
*/
+netdef-forwardIfs[ii].device.pci.domain = virt_fns[ii]-domain;
+netdef-forwardIfs[ii].device.pci.bus = virt_fns[ii]-bus;
+netdef-forwardIfs[ii].device.pci.slot = virt_fns[ii]-slot;
+netdef-forwardIfs[ii].device.pci.function = 
virt_fns[ii]-function;
 }
 netdef-forwardIfs[ii].usageCount = 0;
 }
 
 ret = 0;
 finish:
-for (ii = 0; ii  num_virt_fns; ii++)
+for (ii = 0; ii  num_virt_fns; ii++) {
 VIR_FREE(vfname[ii]);
+VIR_FREE(virt_fns[ii]);
+}
 VIR_FREE(vfname);
+VIR_FREE(virt_fns);
 return ret;
 }
 
diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c
index d53352f..a59012f 100644
--- a/src/util/virnetdev.c
+++ b/src/util/virnetdev.c
@@ -983,18 +983,19 @@ virNetDevSysfsDeviceFile(char **pf_sysfs_device_link, 
const char *ifname,
 int
 virNetDevGetVirtualFunctions(const char *pfname,
  char ***vfname,
+ struct pci_config_address ***virt_fns,
  unsigned int *n_vfname)
 {
 int ret = -1, i;
 char *pf_sysfs_device_link = NULL;
 char *pci_sysfs_device_link = NULL;
-struct pci_config_address **virt_fns;
+//struct pci_config_address **virt_fns;
 char *pciConfigAddr;
 
 if (virNetDevSysfsFile(pf_sysfs_device_link, pfname, device)  0)
 return ret;
 
-if (pciGetVirtualFunctions(pf_sysfs_device_link, virt_fns,
+if (pciGetVirtualFunctions(pf_sysfs_device_link, virt_fns,
n_vfname)  0)
 goto cleanup;
 
@@ -1005,10 +1006,10 @@ virNetDevGetVirtualFunctions(const char *pfname,
 
 for (i = 0; i  *n_vfname; i++)
 {
-if (pciGetDeviceAddrString(virt_fns[i]-domain,
-   virt_fns[i]-bus,
-   virt_fns[i]-slot,
-   virt_fns[i]-function,
+if (pciGetDeviceAddrString((*virt_fns)[i]-domain,
+   (*virt_fns)[i]-bus,
+   (*virt_fns)[i]-slot,
+   (*virt_fns)[i]-function,
pciConfigAddr)  0) {
 virReportSystemError(ENOSYS, %s,
  _(Failed to get PCI Config Address String));
@@ -1021,20 +1022,21 @@ virNetDevGetVirtualFunctions(const char *pfname,
 }
 
 if 

[libvirt] [PATCH 6/6 v3] Forward Mode 'Hostdev' qemu driver implementation

[Shradha Shah]

Signed-off-by: Shradha Shah ss...@solarflare.com
---
 src/qemu/qemu_command.c |   14 ++
 1 files changed, 14 insertions(+), 0 deletions(-)

diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c
index 93c018d..0f6b714 100644
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@@ -5030,6 +5030,20 @@ qemuBuildCommandLine(virConnectPtr conn,
  * code here that adds the newly minted hostdev to the
  * hostdevs array).
  */
+if (qemuAssignDeviceHostdevAlias(def,
+ 
virDomainNetGetActualHostdev(net),
+ (def-nhostdevs-1))  0) {
+qemuReportError(VIR_ERR_INTERNAL_ERROR, %s,
+_(Could not assign alias to Net 
Hostdev));
+goto error;
+}
+
+if (virDomainHostdevInsert(def, 
+   virDomainNetGetActualHostdev(net)) 
 0) {
+qemuReportError(VIR_ERR_INTERNAL_ERROR, %s,
+_(Hostdev not inserted into the array));
+goto error;
+}
 continue;
 }
 
-- 
1.7.4.4

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCH 5/6 v3] Forward Mode Hostdev network driver Implementation

[Shradha Shah]

This patch updates the network driver to properly utilize the new
attributes/elements that are now in virNetworkDef

Signed-off-by: Shradha Shah ss...@solarflare.com
---
 docs/formatnetwork.html.in  |   62 +
 src/network/bridge_driver.c |  213 ---
 2 files changed, 240 insertions(+), 35 deletions(-)

diff --git a/docs/formatnetwork.html.in b/docs/formatnetwork.html.in
index 7e8e991..96b9eb2 100644
--- a/docs/formatnetwork.html.in
+++ b/docs/formatnetwork.html.in
@@ -210,6 +210,37 @@
 (usually either a domain start, or a hotplug interface
 attach to a domain).span class=sinceSince 0.9.4/span
   /dd
+  dtcodehostdev/code/dt
+  dd
+This network facilitates PCI Passthrough of a network device.
+A network device is chosen from the interface pool and 
+directly assigned to the guest using generic device
+passthrough, after first optionally setting the device's MAC
+address to the configured value, and associating the device with
+an 802.1Qbh capable switch using an optionally specified 
+codelt;virtualportgt;/code element. 
+Note that - due to limitations in standard single-port PCI 
+ethernet card driver design - only SR-IOV (Single Root I/O 
+Virtualization) virtual function (VF) devices can be assigned 
+in this manner; to assign a standard single-port PCI or PCIe 
+ethernet card to a guest, use the traditional codelt;
+hostdevgt;/code device definition and span class=since
+Since 0.9.12/span
+
+pNote that this intelligent passthrough of network devices is
+very similar to the functionality of a standard codelt;
+hostdevgt;/code device, the difference being that this 
+method allows specifying a MAC address and codelt;virtualport
+gt;/code for the passed-through device. If these capabilities 
+are not required, if you have a standard single-port PCI, PCIe, 
+or USB network card that doesn't support SR-IOV (and hence would 
+anyway lose the configured MAC address during reset after being 
+assigned to the guest domain), or if you are using a version of 
+libvirt older than 0.9.12, you should use standard 
+codelt;hostdevgt;/code to assign the device to the
+guest instead of codelt;forward mode='hostdev'/gt;/code.
+/p
+  /dd
 /dl
 As mentioned above, a codelt;forwardgt;/code element can
 have multiple codelt;interfacegt;/code subelements, each
@@ -249,6 +280,37 @@
 particular, 'passthrough' mode, and 'private' mode when using
 802.1Qbh), libvirt will choose an unused physical interface
 or, if it can't find an unused interface, fail the operation./p
+
+span class=sincesince 0.9.12/span and when using forward mode 
+'hostdev' we specify the interface pool by using the 
+codelt;addressgt;/code element and codelt;
+typegt;/code codelt;domaingt;/code codelt;busgt;/code
+codelt;slotgt;/code and codelt;functiongt;/code
+sub-elements.
+
+pre
+...
+  lt;forward mode='hostdev' managed='yes'gt;
+lt;address type='pci' domain='0' bus='4' slot='0' function='1'/gt;
+lt;address type='pci' domain='0' bus='4' slot='0' function='2'/gt;
+lt;address type='pci' domain='0' bus='4' slot='0' function='3'/gt;
+  lt;/forwardgt; 
+...
+/pre
+
+Alternatively the interface pool can also be mentioned using a
+single physical function  codelt;pfgt;/code subelement to 
+call out the  corresponding physical interface associated with 
+multiple virtual interfaces (similar to the passthrough mode):
+
+pre
+...
+  lt;forward mode='hostdev' managed='yes'gt;
+lt;pf dev='eth0'/gt;
+  lt;/forwardgt;
+...
+/pre
+
   /dd
 /dl
 h5a name=elementQoSQuality of service/a/h5
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index 2f8a937..c2c763a 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -1938,7 +1938,7 @@ networkStartNetworkExternal(struct network_driver *driver 
ATTRIBUTE_UNUSED,
 virNetworkObjPtr network ATTRIBUTE_UNUSED)
 {
 /* put anything here that needs to be done each time a network of
- * type BRIDGE, PRIVATE, VEPA, or PASSTHROUGH is started. On
+ * type BRIDGE, PRIVATE, VEPA, HOSTDEV or PASSTHROUGH is started. On
  * failure, undo anything you've done, and return -1. On success
  * return 0.
  */
@@ -1949,7 +1949,7 @@ static int networkShutdownNetworkExternal(struct 
network_driver *driver ATTRIBUT
 virNetworkObjPtr network 
ATTRIBUTE_UNUSED)
 {
 /* put anything here that needs to 

Re: [libvirt] availability of release candidate 2 of 0.9.13

[Jim Fehlig]

Daniel Veillard wrote:
I just tagged the git tree and I pushed the tarball for rc2:
  ftp://libvirt.org/libvirt/libvirt-0.9.13-rc2.tar.gz

  the rpms are being built and will be pushed soon too
 I didn't get any feedback on other platforms or OSes for rc1,
 hopefully this will get a bit of testing before the actual
 release, probably on Monday early next week.
   

No problems noted with standard builds in the openSUSE build service

https://build.opensuse.org/package/show?package=libvirtproject=home%3Ajfehlig

Same for SLES builds on the internal build service, which includes more
architectures.

Regards,
Jim

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] qemu_agent: support guest-info command

[Daniel P. Berrange]

On Fri, Jun 29, 2012 at 01:58:05PM +0900, MATSUDA, Daiki wrote:

 diff -uNrp libvirt-0.9.13.orig/daemon/remote.c libvirt-0.9.13/daemon/remote.c
 --- libvirt-0.9.13.orig/daemon/remote.c   2012-06-25 16:06:18.0 
 +0900
 +++ libvirt-0.9.13/daemon/remote.c2012-06-29 12:50:03.752806682 +0900
 @@ -3923,6 +3923,42 @@ cleanup:
  return rv;
  }
  
 +
 +static int
 +remoteDispatchDomainQemuAgentCommand(virNetServerPtr server ATTRIBUTE_UNUSED,
 + virNetServerClientPtr client 
 ATTRIBUTE_UNUSED,
 + virNetMessagePtr msg ATTRIBUTE_UNUSED,
 + virNetMessageErrorPtr rerr,
 + remote_domain_qemu_agent_command_args 
 *args,
 + remote_domain_qemu_agent_command_ret 
 *ret)
 +{
 +virDomainPtr dom = NULL;
 +int rv = -1;
 +struct daemonClientPrivate *priv =
 +virNetServerClientGetPrivateData(client);
 +
 +if (!priv-conn) {
 +virNetError(VIR_ERR_INTERNAL_ERROR, %s, _(connection not open));
 +goto cleanup;
 +}
 +
 +if (!(dom = get_nonnull_domain(priv-conn, args-dom)))
 +goto cleanup;
 +
 +if (virDomainQemuAgentCommand(dom, args-cmd, ret-result, args-flags) 
 0) {
 +virNetError(VIR_ERR_INTERNAL_ERROR, %s, _(Guest Agent Error));
 +goto cleanup;
 +}
 +
 +rv = 0;
 +cleanup:
 +if (rv  0)
 +virNetMessageSaveError(rerr);
 +if (dom)
 +virDomainFree(dom);
 +return rv;
 +}
 +
  /*- Helpers. -*/
  
  /* get_nonnull_domain and get_nonnull_network turn an on-wire
 diff -uNrp libvirt-0.9.13.orig/daemon/remote_dispatch.h 
 libvirt-0.9.13/daemon/remote_dispatch.h
 --- libvirt-0.9.13.orig/daemon/remote_dispatch.h  2012-06-25 
 19:48:08.0 +0900
 +++ libvirt-0.9.13/daemon/remote_dispatch.h   2012-06-29 10:21:21.460454579 
 +0900
 @@ -12889,6 +12889,28 @@ static int remoteDispatchSupportsFeature
  
  
  
 +static int remoteDispatchDomainQemuAgentCommand(
 +virNetServerPtr server,
 +virNetServerClientPtr client,
 +virNetMessagePtr msg,
 +virNetMessageErrorPtr rerr,
 +remote_domain_qemu_agent_command_args *args,
 +remote_domain_qemu_agent_command_ret *ret);
 +static int remoteDispatchDomainQemuAgentCommandHelper(
 +virNetServerPtr server,
 +virNetServerClientPtr client,
 +virNetMessagePtr msg,
 +virNetMessageErrorPtr rerr,
 +void *args,
 +void *ret)
 +{
 +  VIR_DEBUG(server=%p client=%p msg=%p rerr=%p args=%p ret=%p, server, 
 client, msg, rerr, args, ret);
 +  return remoteDispatchDomainQemuAgentCommand(server, client, msg, rerr, 
 args, ret);
 +}
 +/* remoteDispatchDomainQemuAgentCommand body has to be implemented manually 
 */
 +
 +
 +
  virNetServerProgramProc remoteProcs[] = {
  { /* Unused 0 */
 NULL,
 @@ -15374,5 +15396,14 @@ virNetServerProgramProc remoteProcs[] = 
 true,
 1
  },
 +{ /* Method DomainQemuAgentCommand = 276 */
 +   remoteDispatchDomainQemuAgentCommandHelper,
 +   sizeof(remote_domain_qemu_agent_command_args),
 +   (xdrproc_t)xdr_remote_qemu_agent_command_args,
 +   sizeof(remote_domain_qemu_agent_command_ret),
 +   (xdrproc_t)remote_domain_qemu_agent_command_ret,
 +   true,
 +   0
 +},
  };
  size_t remoteNProcs = ARRAY_CARDINALITY(remoteProcs);

This is an auto-generated file. Instead of doing a diff against
two unpacked tar.gz archives, developer against a git checkout,
and use GIT to produce the patch without the auto-generated
cruft.

 diff -uNrp libvirt-0.9.13.orig/include/libvirt/libvirt.h.in 
 libvirt-0.9.13/include/libvirt/libvirt.h.in
 --- libvirt-0.9.13.orig/include/libvirt/libvirt.h.in  2012-06-25 
 21:42:32.0 +0900
 +++ libvirt-0.9.13/include/libvirt/libvirt.h.in   2012-06-29 
 11:22:38.113455058 +0900
 @@ -4132,6 +4132,9 @@ typedef struct _virTypedParameter virMem
   */
  typedef virMemoryParameter *virMemoryParameterPtr;
  
 +int virDomainQemuAgentCommand(virDomainPtr domain, const char *cmd,
 +  char **result, unsigned int flags);
 +
  #ifdef __cplusplus
  }
  #endif


This should be in libvirt-qemu.h



 diff -uNrp libvirt-0.9.13.orig/src/libvirt.c libvirt-0.9.13/src/libvirt.c
 --- libvirt-0.9.13.orig/src/libvirt.c 2012-06-28 12:05:04.0 +0900
 +++ libvirt-0.9.13/src/libvirt.c  2012-06-29 12:51:51.336454508 +0900
 @@ -18973,3 +18973,44 @@ error:
  virDispatchError(dom-conn);
  return -1;
  }
 +
 +/**
 + * virDomainQemuAgentCommand:
 + * @domain: a domain object
 + * @cmd: execution command on domain's guest agent
 + * @result: returning strings
 + * @flags: execution flags
 + *
 + * Provide a list of Guest Agent's support command.
 + * Returns 0 if succeeded, -1 in failing.
 + */
 +int
 +virDomainQemuAgentCommand(virDomainPtr domain,
 +  const char *cmd,
 +  char **result,
 +  unsigned int flags)
 +{
 

[libvirt] [PATCHv3 1/5] qemu: Extended qemuDomainAssignAddresses to be callable from everywhere.

[Viktor Mihajlovski]

This is in preparation of the enablement of s390 guests with virtio devices.

The assignment of device addresses happens in different places, i.e. the
qemu driver and process modules as well as in the unit tests in slightly
different flavors. Currently, these are PPC spapr-vio and PCI
devices, virtio-s390 (not PCI based) will follow.

By optionally passing to qemuDomainAssignAddresses the domain
object and the capabilities it is now possible to call the function
from most of the places (except for hotplug) where address assignment
is done.

Signed-off-by: Viktor Mihajlovski mihaj...@linux.vnet.ibm.com
---
 src/qemu/qemu_command.c |   41 -
 src/qemu/qemu_command.h |6 --
 src/qemu/qemu_driver.c  |   14 +++---
 src/qemu/qemu_process.c |   42 --
 4 files changed, 47 insertions(+), 56 deletions(-)

diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c
index 6549f57..5edf915 100644
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@@ -942,16 +942,22 @@ cleanup:
 
 
 int
-qemuDomainAssignPCIAddresses(virDomainDefPtr def)
+qemuDomainAssignPCIAddresses(virDomainDefPtr def, virBitmapPtr qemuCaps,
+ virDomainObjPtr obj)
 {
 int ret = -1;
-virBitmapPtr qemuCaps = NULL;
+virBitmapPtr localCaps = NULL;
 qemuDomainPCIAddressSetPtr addrs = NULL;
+qemuDomainObjPrivatePtr priv = NULL;
 
-if (qemuCapsExtractVersionInfo(def-emulator, def-os.arch,
-   NULL,
-   qemuCaps)  0)
-goto cleanup;
+if (!qemuCaps) {
+/* need to get information from real environment */
+if (qemuCapsExtractVersionInfo(def-emulator, def-os.arch,
+   NULL,
+   localCaps)  0)
+goto cleanup;
+qemuCaps = localCaps;
+}
 
 if (qemuCapsGet(qemuCaps, QEMU_CAPS_DEVICE)) {
 if (!(addrs = qemuDomainPCIAddressSetCreate(def)))
@@ -961,16 +967,33 @@ qemuDomainAssignPCIAddresses(virDomainDefPtr def)
 goto cleanup;
 }
 
+if (obj  obj-privateData) {
+priv = obj-privateData;
+if (addrs) {
+/* if this is the live domain object, we persist the PCI 
addresses*/
+if (priv-pciaddrs) {
+qemuDomainPCIAddressSetFree(priv-pciaddrs);
+priv-pciaddrs = NULL;
+}
+priv-persistentAddrs = 1;
+priv-pciaddrs = addrs;
+addrs = NULL;
+} else {
+priv-persistentAddrs = 0;
+}
+}
+
 ret = 0;
 
 cleanup:
-qemuCapsFree(qemuCaps);
+qemuCapsFree(localCaps);
 qemuDomainPCIAddressSetFree(addrs);
 
 return ret;
 }
 
-int qemuDomainAssignAddresses(virDomainDefPtr def)
+int qemuDomainAssignAddresses(virDomainDefPtr def, virBitmapPtr qemuCaps,
+  virDomainObjPtr obj)
 {
 int rc;
 
@@ -978,7 +1001,7 @@ int qemuDomainAssignAddresses(virDomainDefPtr def)
 if (rc)
 return rc;
 
-return qemuDomainAssignPCIAddresses(def);
+return qemuDomainAssignPCIAddresses(def, qemuCaps, obj);
 }
 
 static void
diff --git a/src/qemu/qemu_command.h b/src/qemu/qemu_command.h
index 1eafeb3..dd104d6 100644
--- a/src/qemu/qemu_command.h
+++ b/src/qemu/qemu_command.h
@@ -175,10 +175,12 @@ virDomainDefPtr qemuParseCommandLinePid(virCapsPtr caps,
 virDomainChrSourceDefPtr *monConfig,
 bool *monJSON);
 
-int qemuDomainAssignAddresses(virDomainDefPtr def);
+int qemuDomainAssignAddresses(virDomainDefPtr def, virBitmapPtr qemuCaps,
+  virDomainObjPtr);
 int qemuDomainAssignSpaprVIOAddresses(virDomainDefPtr def);
 
-int qemuDomainAssignPCIAddresses(virDomainDefPtr def);
+int qemuDomainAssignPCIAddresses(virDomainDefPtr def, virBitmapPtr qemuCaps,
+ virDomainObjPtr obj);
 qemuDomainPCIAddressSetPtr qemuDomainPCIAddressSetCreate(virDomainDefPtr def);
 int qemuDomainPCIAddressReserveFunction(qemuDomainPCIAddressSetPtr addrs,
 int slot, int function);
diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c
index 2f93404..ef9983c 100644
--- a/src/qemu/qemu_driver.c
+++ b/src/qemu/qemu_driver.c
@@ -1404,7 +1404,7 @@ static virDomainPtr qemudDomainCreate(virConnectPtr conn, 
const char *xml,
 if (qemudCanonicalizeMachine(driver, def)  0)
 goto cleanup;
 
-if (qemuDomainAssignAddresses(def)  0)
+if (qemuDomainAssignAddresses(def, NULL, NULL)  0)
 goto cleanup;
 
 if (!(vm = virDomainAssignDef(driver-caps,
@@ -5070,7 +5070,7 @@ static virDomainPtr qemudDomainDefine(virConnectPtr conn, 
const char *xml) {
 if (qemudCanonicalizeMachine(driver, def)  0)
 goto cleanup;
 
-if (qemuDomainAssignAddresses(def)  0)
+if 

[libvirt] [PATCHv3 0/5] Virtio support for S390

[Viktor Mihajlovski]

As 0.9.13 is stabilizing and I will not be available next week
I am sending this reworked patch set already today, looking forward
to comments.

This series adds support for the s390 flavor of virtio devices.
Since the s390 virtio devices are not implemented as PCI devices
it is necessary to refactor some of the device address assignment
code.

v2 changes
   resent as thread

v3 changes
   renumbered new virtio-s390 capability
   fixed incorrect whitespace
   fixed subject lines

Viktor Mihajlovski (5):
  qemu: Extended qemuDomainAssignAddresses to be callable from
everywhere.
  qemu: Change tests to use (modified) qemuDomainAssignAddresses
  S390: Add support for virtio-s390 devices.
  S390: Domain Schema for s390-virtio machines.
  S390: Adding testcases for s390

 docs/schemas/domaincommon.rng  |   20 +++
 src/conf/domain_conf.c |   11 +-
 src/conf/domain_conf.h |1 +
 src/qemu/qemu_capabilities.c   |7 +
 src/qemu/qemu_capabilities.h   |1 +
 src/qemu/qemu_command.c|  139 ++--
 src/qemu/qemu_command.h|6 +-
 src/qemu/qemu_driver.c |   14 +-
 src/qemu/qemu_process.c|   42 +--
 .../qemuxml2argv-console-virtio-s390.args  |9 ++
 .../qemuxml2argv-console-virtio-s390.xml   |   24 
 .../qemuxml2argv-disk-virtio-s390.args |5 +
 .../qemuxml2argv-disk-virtio-s390.xml  |   22 +++
 .../qemuxml2argv-minimal-s390.args |5 +
 .../qemuxml2argvdata/qemuxml2argv-minimal-s390.xml |   21 +++
 .../qemuxml2argv-net-virtio-s390.args  |5 +
 .../qemuxml2argv-net-virtio-s390.xml   |   22 +++
 tests/qemuxml2argvtest.c   |   20 ++--
 tests/qemuxmlnstest.c  |   13 +--
 tests/testutilsqemu.c  |   31 +
 20 files changed, 332 insertions(+), 86 deletions(-)
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.xml
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.xml
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.xml
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-s390.xml

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCHv3 2/5] qemu: Change tests to use (modified) qemuDomainAssignAddresses

[Viktor Mihajlovski]

Rewrote the device assignment parts in tests to use qemuDomainAssignAddresses.
This way the tests will work for new device address types as they show
up in the future (like s390 device types).

Signed-off-by: Viktor Mihajlovski mihaj...@linux.vnet.ibm.com
---
 tests/qemuxml2argvtest.c |   12 +---
 tests/qemuxmlnstest.c|   13 ++---
 2 files changed, 3 insertions(+), 22 deletions(-)

diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c
index 7b00ea2..cda32b6 100644
--- a/tests/qemuxml2argvtest.c
+++ b/tests/qemuxml2argvtest.c
@@ -149,21 +149,11 @@ static int testCompareXMLToArgvFiles(const char *xml,
 goto out;
 
 if (qemuCapsGet(extraFlags, QEMU_CAPS_DEVICE)) {
-qemuDomainPCIAddressSetPtr pciaddrs;
-
-if (qemuDomainAssignSpaprVIOAddresses(vmdef)) {
+if (qemuDomainAssignAddresses(vmdef, extraFlags, NULL)) {
 if (expectError)
 goto ok;
 goto out;
 }
-
-if (!(pciaddrs = qemuDomainPCIAddressSetCreate(vmdef)))
-goto out;
-
-if (qemuAssignDevicePCISlots(vmdef, pciaddrs)  0)
-goto out;
-
-qemuDomainPCIAddressSetFree(pciaddrs);
 }
 
 log = virtTestLogContentAndReset();
diff --git a/tests/qemuxmlnstest.c b/tests/qemuxmlnstest.c
index 8eca466..0bc821d 100644
--- a/tests/qemuxmlnstest.c
+++ b/tests/qemuxmlnstest.c
@@ -95,17 +95,8 @@ static int testCompareXMLToArgvFiles(const char *xml,
 if (qemudCanonicalizeMachine(driver, vmdef)  0)
 goto fail;
 
-if (qemuCapsGet(extraFlags, QEMU_CAPS_DEVICE)) {
-qemuDomainPCIAddressSetPtr pciaddrs;
-if (!(pciaddrs = qemuDomainPCIAddressSetCreate(vmdef)))
-goto fail;
-
-if (qemuAssignDevicePCISlots(vmdef, pciaddrs)  0)
-goto fail;
-
-qemuDomainPCIAddressSetFree(pciaddrs);
-}
-
+if (qemuCapsGet(extraFlags, QEMU_CAPS_DEVICE))
+qemuDomainAssignAddresses(vmdef, extraFlags, NULL);
 
 log = virtTestLogContentAndReset();
 VIR_FREE(log);
-- 
1.7.0.4

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCHv3 4/5] S390: Domain Schema for s390-virtio machines.

[Viktor Mihajlovski]

Added s390-virtio machine type to the XML schema for domains in order
to not fail the domain schema tests.

Signed-off-by: Viktor Mihajlovski mihaj...@linux.vnet.ibm.com
---
 docs/schemas/domaincommon.rng |   20 
 1 files changed, 20 insertions(+), 0 deletions(-)

diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng
index 912a1a2..70c7d16 100644
--- a/docs/schemas/domaincommon.rng
+++ b/docs/schemas/domaincommon.rng
@@ -283,6 +283,7 @@
   ref name=hvmsparc/
   ref name=hvmppc/
   ref name=hvmppc64/
+  ref name=hvms390/
 /choice
   /optional
   valuehvm/value
@@ -369,6 +370,25 @@
   /optional
 /group
   /define
+  define name=hvms390
+group
+  optional
+attribute name=arch
+  choice
+values390/value
+values390x/value
+  /choice
+/attribute
+  /optional
+  optional
+attribute name=machine
+  choice
+values390-virtio/value
+  /choice
+/attribute
+  /optional
+/group
+  /define
   define name=osexe
 element name=os
   element name=type
-- 
1.7.0.4

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCHv3 5/5] S390: Adding testcases for s390

[Viktor Mihajlovski]

Add minimal s390-virtio domain testcase and testcases for virtio serial,
net, disk for the virtio-s390 bus.

Signed-off-by: Viktor Mihajlovski mihaj...@linux.vnet.ibm.com
---
 .../qemuxml2argv-console-virtio-s390.args  |9 ++
 .../qemuxml2argv-console-virtio-s390.xml   |   24 +++
 .../qemuxml2argv-disk-virtio-s390.args |5 +++
 .../qemuxml2argv-disk-virtio-s390.xml  |   22 ++
 .../qemuxml2argv-minimal-s390.args |5 +++
 .../qemuxml2argvdata/qemuxml2argv-minimal-s390.xml |   21 +
 .../qemuxml2argv-net-virtio-s390.args  |5 +++
 .../qemuxml2argv-net-virtio-s390.xml   |   22 ++
 tests/qemuxml2argvtest.c   |8 +
 tests/testutilsqemu.c  |   31 
 10 files changed, 152 insertions(+), 0 deletions(-)
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.xml
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.xml
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.xml
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-s390.args
 create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-net-virtio-s390.xml

diff --git a/tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.args 
b/tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.args
new file mode 100644
index 000..3388a35
--- /dev/null
+++ b/tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.args
@@ -0,0 +1,9 @@
+LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu -S -M \
+s390-virtio -m 214 -smp 1 -nographic -nodefconfig -nodefaults -chardev \
+socket,id=charmonitor,path=/tmp/test-monitor,server,nowait -mon \
+chardev=charmonitor,id=monitor,mode=readline -no-acpi \
+-boot c -device virtio-serial-s390,id=virtio-serial0 \
+-drive file=/dev/HostVG/QEMUGuest1,if=none,id=drive-virtio-disk0 \
+-device virtio-blk-s390,drive=drive-virtio-disk0,id=virtio-disk0 \
+-chardev pty,id=charconsole0 \
+-device virtconsole,chardev=charconsole0,id=console0 -usb
diff --git a/tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.xml 
b/tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.xml
new file mode 100644
index 000..5a4a9d4
--- /dev/null
+++ b/tests/qemuxml2argvdata/qemuxml2argv-console-virtio-s390.xml
@@ -0,0 +1,24 @@
+domain type='qemu'
+  nameQEMUGuest1/name
+  uuidc7a5fdbd-edaf-9455-926a-d65c16db1809/uuid
+  memory219100/memory
+  currentMemory219100/currentMemory
+  os
+type arch='s390x' machine='s390-virtio'hvm/type
+boot dev='hd'/
+  /os
+  clock offset='utc'/
+  on_poweroffdestroy/on_poweroff
+  on_rebootrestart/on_reboot
+  on_crashdestroy/on_crash
+  devices
+emulator/usr/bin/qemu/emulator
+disk type='block' device='disk'
+  source dev='/dev/HostVG/QEMUGuest1'/
+  target dev='hda' bus='virtio'/
+/disk
+console type='pty'
+  target type='virtio'/
+/console
+  /devices
+/domain
diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.args 
b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.args
new file mode 100644
index 000..66bdf1e
--- /dev/null
+++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.args
@@ -0,0 +1,5 @@
+LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test /usr/bin/qemu -S \
+-M s390-virtio -m 214 -smp 1 -nographic -nodefaults \
+-monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c \
+-drive file=/dev/HostVG/QEMUGuest1,if=none,id=drive-virtio-disk0 \
+-device virtio-blk-s390,drive=drive-virtio-disk0,id=virtio-disk0 -usb
diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.xml 
b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.xml
new file mode 100644
index 000..9ce6c18
--- /dev/null
+++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-s390.xml
@@ -0,0 +1,22 @@
+domain type='qemu'
+  nameQEMUGuest1/name
+  uuidc7a5fdbd-edaf-9455-926a-d65c16db1809/uuid
+  memory219136/memory
+  currentMemory219136/currentMemory
+  vcpu1/vcpu
+  os
+type arch='s390x' machine='s390-virtio'hvm/type
+boot dev='hd'/
+  /os
+  clock offset='utc'/
+  on_poweroffdestroy/on_poweroff
+  on_rebootrestart/on_reboot
+  on_crashdestroy/on_crash
+  devices
+emulator/usr/bin/qemu/emulator
+disk type='block' device='disk'
+  source dev='/dev/HostVG/QEMUGuest1'/
+  target dev='hda' bus='virtio'/
+/disk
+  /devices
+/domain
diff --git a/tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.args 
b/tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.args
new file mode 100644
index 000..4e7bf05
--- /dev/null
+++ b/tests/qemuxml2argvdata/qemuxml2argv-minimal-s390.args
@@ -0,0 +1,5 @@
+LC_ALL=C PATH=/bin 

[libvirt] [PATCHv3 3/5] S390: Add support for virtio-s390 devices.

[Viktor Mihajlovski]

The s390(x) architecture doesn't feature a PCI bus. For the purpose of
supporting virtio devices a virtual bus called virtio-s390 is used.
A new address type VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_S390 is used to
distinguish the virtio devices on s390 from PCI-based virtio devices.

V3 Change: updated QEMU_CAPS_VIRTIO_S390 to fit upstream.

Signed-off-by: Viktor Mihajlovski mihaj...@linux.vnet.ibm.com
---
 src/conf/domain_conf.c   |   11 +++-
 src/conf/domain_conf.h   |1 +
 src/qemu/qemu_capabilities.c |7 +++
 src/qemu/qemu_capabilities.h |1 +
 src/qemu/qemu_command.c  |   98 +++--
 5 files changed, 110 insertions(+), 8 deletions(-)

diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index 4086dac..cf7c757 100644
--- a/src/conf/domain_conf.c
+++ b/src/conf/domain_conf.c
@@ -149,7 +149,8 @@ VIR_ENUM_IMPL(virDomainDeviceAddress, 
VIR_DOMAIN_DEVICE_ADDRESS_TYPE_LAST,
   virtio-serial,
   ccid,
   usb,
-  spapr-vio)
+  spapr-vio,
+  virtio-s390)
 
 VIR_ENUM_IMPL(virDomainDeviceAddressPciMulti,
   VIR_DOMAIN_DEVICE_ADDRESS_PCI_MULTI_LAST,
@@ -2132,7 +2133,8 @@ virDomainDeviceInfoFormat(virBufferPtr buf,
 virBufferAddLit(buf, /\n);
 }
 
-if (info-type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE)
+if (info-type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE ||
+info-type == VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_S390)
 return 0;
 
 /* We'll be in domain/devices/[device type]/ so 3 level indent */
@@ -4123,6 +4125,7 @@ virDomainControllerDefParseXML(xmlNodePtr node,
 
 if (def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE 
 def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_SPAPRVIO 
+def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_S390 
 def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) {
 virDomainReportError(VIR_ERR_INTERNAL_ERROR, %s,
  _(Controllers must use the 'pci' address type));
@@ -4676,6 +4679,7 @@ virDomainNetDefParseXML(virCapsPtr caps,
  * them we should make sure address type is correct */
 if (def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE 
 def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_SPAPRVIO 
+def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_S390 
 def-info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) {
 virDomainReportError(VIR_ERR_INTERNAL_ERROR, %s,
  _(Network interfaces must use 'pci' address 
type));
@@ -9078,7 +9082,8 @@ static virDomainDefPtr virDomainDefParseXML(virCapsPtr 
caps,
 
 def-memballoon = memballoon;
 VIR_FREE(nodes);
-} else {
+} else if (!STREQ(def-os.arch,s390x)) {
+/* TODO: currently no balloon support on s390 - no default balloon */
 if (def-virtType == VIR_DOMAIN_VIRT_XEN ||
 def-virtType == VIR_DOMAIN_VIRT_QEMU ||
 def-virtType == VIR_DOMAIN_VIRT_KQEMU ||
diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h
index 7d5d60b..5e5374a 100644
--- a/src/conf/domain_conf.h
+++ b/src/conf/domain_conf.h
@@ -172,6 +172,7 @@ enum virDomainDeviceAddressType {
 VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCID,
 VIR_DOMAIN_DEVICE_ADDRESS_TYPE_USB,
 VIR_DOMAIN_DEVICE_ADDRESS_TYPE_SPAPRVIO,
+VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_S390,
 
 VIR_DOMAIN_DEVICE_ADDRESS_TYPE_LAST
 };
diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
index 1e12a39..b6e5bd1 100644
--- a/src/qemu/qemu_capabilities.c
+++ b/src/qemu/qemu_capabilities.c
@@ -166,6 +166,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST,
   hda-micro, /* 95 */
   dump-guest-memory,
   nec-usb-xhci,
+  virtio-s390,
 
 );
 
@@ -1430,6 +1431,12 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr 
flags)
 qemuCapsSet(flags, QEMU_CAPS_USB_HUB);
 if (strstr(str, name \ich9-ahci\))
 qemuCapsSet(flags, QEMU_CAPS_ICH9_AHCI);
+if (strstr(str, name \virtio-blk-s390\))
+qemuCapsSet(flags, QEMU_CAPS_VIRTIO_S390);
+if (strstr(str, name \virtio-net-s390\))
+qemuCapsSet(flags, QEMU_CAPS_VIRTIO_S390);
+if (strstr(str, name \virtio-serial-s390\))
+qemuCapsSet(flags, QEMU_CAPS_VIRTIO_S390);
 
 /* Prefer -chardev spicevmc (detected earlier) over -device spicevmc */
 if (!qemuCapsGet(flags, QEMU_CAPS_CHARDEV_SPICEVMC) 
diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h
index 83c135b..9b5ff30 100644
--- a/src/qemu/qemu_capabilities.h
+++ b/src/qemu/qemu_capabilities.h
@@ -133,6 +133,7 @@ enum qemuCapsFlags {
 QEMU_CAPS_HDA_MICRO  = 95, /* -device hda-micro */
 QEMU_CAPS_DUMP_GUEST_MEMORY  = 96, /* dump-guest-memory command */
 QEMU_CAPS_NEC_USB_XHCI   = 97, /* -device nec-usb-xhci */
+QEMU_CAPS_VIRTIO_S390= 98, /* -device virtio-*-s390 */
 
 

[libvirt] [PATCH] Remove sub-mounts under /dev when starting an LXC container

[Daniel P. Berrange]

From: Daniel P. Berrange berra...@redhat.com

Since we are mounting a new /dev in the container, we must
remove any sub-mounts like /dev/shm, /dev/mqueue, etc,
otherwise they'll be recorded in /proc/mounts, but not be
accessible to applications.
---
 src/lxc/lxc_container.c |7 ---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c
index 071d8d1..910e82b 100644
--- a/src/lxc/lxc_container.c
+++ b/src/lxc/lxc_container.c
@@ -1416,10 +1416,11 @@ static int lxcContainerSetupPivotRoot(virDomainDefPtr 
vmDef,
 if (lxcContainerPivotRoot(root)  0)
 goto cleanup;
 
-/* Gets rid of any existing stuff under /proc, since we need new
- * namespace aware versions of those. We must do /proc second
- * otherwise we won't find /proc/mounts :-) */
+/* Gets rid of any existing stuff under /proc, /sys  /tmp
+ * We need new namespace aware versions of those. We must
+ * do /proc last otherwise we won't find /proc/mounts :-) */
 if (lxcContainerUnmountSubtree(/sys, false)  0 ||
+lxcContainerUnmountSubtree(/dev, false)  0 ||
 lxcContainerUnmountSubtree(/proc, false)  0)
 goto cleanup;
 
-- 
1.7.10.4

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] Different mac address used in libvirtd

[Daniel P. Berrange]

On Wed, Jun 20, 2012 at 07:01:17PM +0800, Yong Sheng Gong wrote:
 Hi,
 I found libvirtd will create vnet0 with a different mac other than the one 
 defined in domain file. The mac in Domain file will be used by qemu-kvm.
 I want to know how libvirtd decides the mac for vnet0, Who can give me a hint 
 about the location of the libvirtd codes which does this work?
 
 I cannot find how the libvirtd translates the mac from one defined in domain 
 file to vnet0's one in src/util/bridge.c: brSetInterfaceMac()

     interface type='bridge'
   mac address='fa:16:3e:28:64:c2'/
   source bridge='br100'/
   target dev='vnet0'/
   filterref filter='nova-instance-instance-0002-fa163e2864c2'
     parameter name='DHCPSERVER' value='10.0.1.1'/
     parameter name='PROJNET' value='10.0.1.0'/
     parameter name='PROJMASK' value='255.255.255.0'/
     parameter name='IP' value='10.0.1.2'/
   /filterref
   alias name='net0'/
   address type='pci' domain='0x' bus='0x00' slot='0x03' 
 function='0x0'/
     /interface

The XML mac address config, sets the MAC address for the *guest*
NIC, ie what the guest uses as its eth0.



 [root@robinlinux eclipsecdt]# ifconfig  vnet0
 vnet0 Link encap:Ethernet  HWaddr FE:16:3E:28:64:C2  
   inet6 addr: fe80::fc16:3eff:fe28:64c2/64 Scope:Link
   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
   RX packets:553 errors:0 dropped:0 overruns:0 frame:0
   TX packets:1485 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:500 
   RX bytes:136360 (133.1 KiB)  TX bytes:280590 (274.0 KiB)

vnet0 is the backend of the guest NIC, and its MAC addr
is more or less irrelevant to functioning of the guest
itself, since traffic does not originate on this NIC.
The only important thing is that this TAP device must
have a high value MAC address, to avoid the bridge
device using the TAP device's MAC as its own. Hence
when creating the TAP Device  libvirt takes the guest
MAC addr and simply sets the top byte to 0xFE


Daniel
-- 
|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org  -o- http://virt-manager.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] Remove sub-mounts under /dev when starting an LXC container

[Eric Blake]

On 06/29/2012 09:30 AM, Daniel P. Berrange wrote:
 From: Daniel P. Berrange berra...@redhat.com
 
 Since we are mounting a new /dev in the container, we must
 remove any sub-mounts like /dev/shm, /dev/mqueue, etc,
 otherwise they'll be recorded in /proc/mounts, but not be
 accessible to applications.
 ---
  src/lxc/lxc_container.c |7 ---
  1 file changed, 4 insertions(+), 3 deletions(-)

ACK.

-- 
Eric Blake   ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org





signature.asc
Description: OpenPGP digital signature
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] build: define WITH_INTERFACE for the driver

[Eric Blake]

On 06/29/2012 01:34 AM, Osier Yang wrote:
 On 2012年06月29日 07:57, Eric Blake wrote:
 Our code was mistakenly relying on an undefined macro, WITH_INTERFACE,
 for determining whether to load the interface driver which wraps the
 netcf library.  Clean this situation up by having only one automake
 conditional for the driver, and having both WITH_NETCF (library
 detected) and WITH_INTERFACE (driver enabled) in C code, in case a
 future patch ever adds a network management via means other than

s/network/interface/

 the netcf library.
 
 Foresighted. :-)

Trying to model after the storage driver, and how it has several backends.


 -dnl netcf library
 +dnl check if the interface driver should be compiled
 +
 +AC_ARG_WITH([interface],
 +  AC_HELP_STRING([--with-interface],
 +[with host interface driver @:@default=check@:@]),[],
 +[with_interface=check])
 
 Do we have to expose with-interface? It will give the user
 a logic question, pick with-interface, or 'with-netcf', or
 both, even more when we have other implementations of interface
 driver in future. however, the logic is simple, and we do it
 inside actually: as long as one implementation of the interface
 driver is picked to compile, we have the WITH_INTERFACE. so IMHO
 no need to give the user the simple logic question. :-)

Good point.  Looking at how storage did it, we have:

--with-storage-dir
--with-storage-fs
...

but no top-level --with-storage.  That is, you get WITH_STORAGE if any
of the --with-storage-backends ended up as yes.

At first, I was worried about back-compat (old builds were used to
--with-netcf, and I didn't want to break that), but the more I think
about it, the more I think that it's okay to break naming conventions
for something that is easier to explain.

I see two possible solutions, then:

1. Assume that like the storage driver, the interface driver will
eventually have multiple backends.  Then we would have:

--with-interface-netcf

as a way to select the netcf backend in the interface driver, and
WITH_INTERFACE would be automatic if at least one backend (in this case,
netcf being the only backend) is found.

2. Save the complexity of multiple backends for the day when we actually
have multiple backends, and for now just have a single configure option
--with-interface.

Either way, I would completely ditch --with-netcf, and refactor the
logic to be:

if test $with_libvirtd = no; then
with_interface_netcf=no
fi
if test $with_interface_netcf = yes || \
test $with_interface_netcf = check; then
probe for netcf, fail if it was required
fi
if test $with_interface_netcf = yes; then
set WITH_INTERFACE witness
fi

I'll go ahead and respin this patch along those lines.

 And above changes, what we need is just:
 
 if test $with_netfs = yes || with_something_else = yes; then
 AC_DEFINE_UNQUOTED([WITH_INTERFACE], [1],
   [whether interface driver is enabled])
 fi
 AM_CONDITIONAL([WITH_INTERFACE], [test $with_interface = yes])

Yep, we're thinking on the same lines - probe for each backend, then
make the driver decision based on the result of the backend probes, but
only expose the backends as the configure options.


 @@ -2807,11 +2833,12 @@ AC_MSG_NOTICE([ ESX: $with_esx])
   AC_MSG_NOTICE([ Hyper-V: $with_hyperv])
   AC_MSG_NOTICE([Test: $with_test])
   AC_MSG_NOTICE([  Remote: $with_remote])
 -AC_MSG_NOTICE([ Network: $with_network])
   AC_MSG_NOTICE([Libvirtd: $with_libvirtd])
 -AC_MSG_NOTICE([   netcf: $with_netcf])
 
 And no AC_MSG_NOTICE for $with_interface here, with keeping
 $with_netcf.
 
 -AC_MSG_NOTICE([ macvtap: $with_macvtap])
 -AC_MSG_NOTICE([virtport: $with_virtualport])
 +AC_MSG_NOTICE([ Network: $with_network])
 +AC_MSG_NOTICE([   Iface: $with_interface])
 +AC_MSG_NOTICE([ Secrets: $with_secrets])
 +AC_MSG_NOTICE([ NodeDev: $with_nodedev])
 +AC_MSG_NOTICE([NWfilter: $with_nwfilter])

Actually, no with_netcf here (this is the driver section, but with_netcf
is already present in the library section), so we DO want a listing here
of whether the with_interface driver was selected (whether by with_netcf
or by some other backend).


 -%if %{with_netcf}
 -%define with_interface 1
 -%else
 +%if !%{with_netcf}
   %define with_interface 0
   %endif

The logic here would be a bit different; just as the spec file has to
know when to package the storage driver (if any of the storage backends
were selected), we still have to package the interface driver, so this
variable is still useful.


 @@ -1056,6 +1056,9 @@ of recent versions of Linux (and other OSes).
   %define _without_network --without-network
   %endif

 +%if ! %{with_interface}
 +%define _without_interface --without-interface
 +%endif
   %if ! %{with_storage_fs}
   %define _without_storage_fs --without-storage-fs
   %endif
 @@ -1171,6 +1174,7 @@ autoreconf -if
  %{?_without_hyperv} \
  %{?_without_vmware} \
  %{?_without_network} \
 +   %{?_without_interface} \
  

Re: [libvirt] [PATCH] build: define WITH_INTERFACE for the driver

[Daniel P. Berrange]

On Fri, Jun 29, 2012 at 10:18:47AM -0600, Eric Blake wrote:
 On 06/29/2012 01:34 AM, Osier Yang wrote:
  On 2012年06月29日 07:57, Eric Blake wrote:
  Our code was mistakenly relying on an undefined macro, WITH_INTERFACE,
  for determining whether to load the interface driver which wraps the
  netcf library.  Clean this situation up by having only one automake
  conditional for the driver, and having both WITH_NETCF (library
  detected) and WITH_INTERFACE (driver enabled) in C code, in case a
  future patch ever adds a network management via means other than
 
 s/network/interface/
 
  the netcf library.
  
  Foresighted. :-)
 
 Trying to model after the storage driver, and how it has several backends.
 
 
  -dnl netcf library
  +dnl check if the interface driver should be compiled
  +
  +AC_ARG_WITH([interface],
  +  AC_HELP_STRING([--with-interface],
  +[with host interface driver @:@default=check@:@]),[],
  +[with_interface=check])
  
  Do we have to expose with-interface? It will give the user
  a logic question, pick with-interface, or 'with-netcf', or
  both, even more when we have other implementations of interface
  driver in future. however, the logic is simple, and we do it
  inside actually: as long as one implementation of the interface
  driver is picked to compile, we have the WITH_INTERFACE. so IMHO
  no need to give the user the simple logic question. :-)
 
 Good point.  Looking at how storage did it, we have:
 
 --with-storage-dir
 --with-storage-fs
 ...
 
 but no top-level --with-storage.  That is, you get WITH_STORAGE if any
 of the --with-storage-backends ended up as yes.
 
 At first, I was worried about back-compat (old builds were used to
 --with-netcf, and I didn't want to break that), but the more I think
 about it, the more I think that it's okay to break naming conventions
 for something that is easier to explain.
 
 I see two possible solutions, then:
 
 1. Assume that like the storage driver, the interface driver will
 eventually have multiple backends.  Then we would have:
 
 --with-interface-netcf
 
 as a way to select the netcf backend in the interface driver, and
 WITH_INTERFACE would be automatic if at least one backend (in this case,
 netcf being the only backend) is found.
 
 2. Save the complexity of multiple backends for the day when we actually
 have multiple backends, and for now just have a single configure option
 --with-interface.
 
 Either way, I would completely ditch --with-netcf, and refactor the
 logic to be:
 
 if test $with_libvirtd = no; then
 with_interface_netcf=no
 fi
 if test $with_interface_netcf = yes || \
 test $with_interface_netcf = check; then
 probe for netcf, fail if it was required
 fi
 if test $with_interface_netcf = yes; then
 set WITH_INTERFACE witness
 fi
 
 I'll go ahead and respin this patch along those lines.

I'm not a fan of this, because you are too tightly associating use
of the netcf library, with use of the interface drivers, and also
presuming a 1-1 relationship between a logical driver, and an external
library. THis breaks down if a module like the inteface driver needs
to check for multiple external libraries, and if the external libraries
are used by multiple different areas of the libvirt code.

My view is that in the configure script, we have two types of checks
and we must keep them strictly separated.

  - External modules (netcf, lvm, other libraries)
  - Logical modules  (storage driver, network driver, interface driver)

We should first do checks for the external modules. These checks
can be disabled/forced using --with-netcf/--without-netcf

The checks for logical modules, should just look to see if
their all of their prerequisites are present, but again allow
you to turn off the module using  --with-interface/--without-interface


My long term vision is that we one day refactor our enourmous
configure script into a set of isolated modules.

So, you'd be able to declare logical modules

   AC_DEFUN(LIBVIRT_LIBRARY_NETCF, [
 ...code to check for netcf and
CLI args to enable/disable
   ])


   AC_DEFUN(LIBVIRT_DRIVER_INTERFACE,
   [
  AC_REQUIRE([LIBVIRT_DEP_NETCF])

  ...code to enable interface
 driver if netcf was present
   ])

   AC_DEFUN(LIBVIRT_DRIVER_STORAGE,
   [
  AC_REQUIRE([LIBVIRT_DEP_LVM])
  AC_REQUIRE([LIBVIRT_DEP_QEMU_IMG])
  AC_REQUIRE([LIBVIRT_DEP_ISCSI])

  ...code to enable storage
 driver parts...
   ])

and each of these definitions be completely separate .m4 files. So the
eventual libvirt configure.ac script would just be doing

   LIBVIRT_DRIVER_INTERFACE
   LIBVIRT_DRIVER_STORAGE

and so on.

Daniel
-- 
|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org  -o- http://virt-manager.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|

--
libvir-list 

[libvirt] [Patch v2 2/3] Add -netdev bridge support

[rmarwah]

From: Richa Marwaha rmar...@linux.vnet.ibm.com

This patch adds the support to run the QEMU network helper
under unprivileged user. It also adds the support for
attach-interface option in virsh to run under unprivileged
user.

Signed-off-by: Richa Marwaha rmar...@linux.vnet.ibm.com
Signed-off-by: Corey Bryantcor...@linux.vnet.ibm.com
---
v2
- This patch attach-interface option is tested on
commit cd15303fd123146b0ba53e387d08ef22b707223

 src/qemu/qemu_command.c |   61 +-
 src/qemu/qemu_command.h |2 +
 src/qemu/qemu_hotplug.c |   31 ---
 3 files changed, 67 insertions(+), 27 deletions(-)

diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c
index 6549f57..4eb8cd5 100644
--- a/src/qemu/qemu_command.c
+++ b/src/qemu/qemu_command.c
@@ -2851,6 +2851,8 @@ error:
 
 char *
 qemuBuildHostNetStr(virDomainNetDefPtr net,
+struct qemud_driver *driver,
+virBitmapPtr qemuCaps,
 char type_sep,
 int vlan,
 const char *tapfd,
@@ -2859,6 +2861,7 @@ qemuBuildHostNetStr(virDomainNetDefPtr net,
 bool is_tap = false;
 virBuffer buf = VIR_BUFFER_INITIALIZER;
 enum virDomainNetType netType = virDomainNetGetActualType(net);
+const char *brname = NULL;
 
 if (net-script  netType != VIR_DOMAIN_NET_TYPE_ETHERNET) {
 qemuReportError(VIR_ERR_CONFIG_UNSUPPORTED,
@@ -2868,8 +2871,21 @@ qemuBuildHostNetStr(virDomainNetDefPtr net,
 }
 
 switch (netType) {
-case VIR_DOMAIN_NET_TYPE_NETWORK:
+/*
+ * If type='bridge', and we're running as privileged user
+ * or -netdev bridge is not supported then it will fall
+ * through, -net tap,fd
+ */
 case VIR_DOMAIN_NET_TYPE_BRIDGE:
+if (!driver-privileged 
+qemuCapsGet(qemuCaps, QEMU_CAPS_NETDEV_BRIDGE)) {
+brname = virDomainNetGetActualBridgeName(net);
+virBufferAsprintf(buf, bridge%cbr=%s, type_sep, brname);
+type_sep = ',';
+is_tap = true;
+break;
+}
+case VIR_DOMAIN_NET_TYPE_NETWORK:
 case VIR_DOMAIN_NET_TYPE_DIRECT:
 virBufferAsprintf(buf, tap%cfd=%s, type_sep, tapfd);
 type_sep = ',';
@@ -4997,7 +5013,7 @@ qemuBuildCommandLine(virConnectPtr conn,
 for (i = 0 ; i  def-nnets ; i++) {
 virDomainNetDefPtr net = def-nets[i];
 char *nic, *host;
-char tapfd_name[50];
+char tapfd_name[50] = ;
 char vhostfd_name[50] = ;
 int vlan;
 int bootindex = bootNet;
@@ -5034,17 +5050,26 @@ qemuBuildCommandLine(virConnectPtr conn,
 
 if (actualType == VIR_DOMAIN_NET_TYPE_NETWORK ||
 actualType == VIR_DOMAIN_NET_TYPE_BRIDGE) {
-int tapfd = qemuNetworkIfaceConnect(def, conn, driver, net,
-qemuCaps);
-if (tapfd  0)
-goto error;
-
-last_good_net = i;
-virCommandTransferFD(cmd, tapfd);
-
-if (snprintf(tapfd_name, sizeof(tapfd_name), %d,
- tapfd) = sizeof(tapfd_name))
-goto no_memory;
+/*
+ * If type='bridge' then we attempt to allocate the tap fd 
here only if
+ * running under a privilged user or -netdev bridge option is 
not
+ * supported.
+ */
+ if (actualType == VIR_DOMAIN_NET_TYPE_NETWORK ||
+ driver-privileged ||
+ (!qemuCapsGet(qemuCaps, QEMU_CAPS_NETDEV_BRIDGE))) {
+ int tapfd = qemuNetworkIfaceConnect(def, conn, driver, 
net,
+ qemuCaps);
+ if (tapfd  0)
+ goto error;
+
+ last_good_net = i;
+ virCommandTransferFD(cmd, tapfd);
+
+  if (snprintf(tapfd_name, sizeof(tapfd_name), %d,
+   tapfd) = sizeof(tapfd_name))
+  goto no_memory;
+ }
 } else if (actualType == VIR_DOMAIN_NET_TYPE_DIRECT) {
 int tapfd = qemuPhysIfaceConnect(def, driver, net,
  qemuCaps, vmop);
@@ -5087,8 +5112,9 @@ qemuBuildCommandLine(virConnectPtr conn,
 if (qemuCapsGet(qemuCaps, QEMU_CAPS_NETDEV) 
 qemuCapsGet(qemuCaps, QEMU_CAPS_DEVICE)) {
 virCommandAddArg(cmd, -netdev);
-if (!(host = qemuBuildHostNetStr(net, ',', vlan,
- tapfd_name, vhostfd_name)))
+if (!(host = qemuBuildHostNetStr(net, driver, qemuCaps,
+ ',', vlan, tapfd_name,
+  

[libvirt] [Patch v2 3/3] apparmor: QEMU bridge helper policy updates

[rmarwah]

From: Richa Marwaha rmar...@linux.vnet.ibm.com

This patch provides AppArmor policy updates for the QEMU bridge helper.
The QEMU bridge helper is a SUID executable exec'd by QEMU that drops
capabilities to CAP_NET_ADMIN and adds a tap device to a network bridge.

Signed-off-by: Richa Marwaha rmar...@linux.vnet.ibm.com
Signed-off-by: Corey Bryantcor...@linux.vnet.ibm.com
---
 examples/apparmor/libvirt-qemu |   21 -
 1 files changed, 20 insertions(+), 1 deletions(-)

diff --git a/examples/apparmor/libvirt-qemu b/examples/apparmor/libvirt-qemu
index 10cdd36..766a334 100644
--- a/examples/apparmor/libvirt-qemu
+++ b/examples/apparmor/libvirt-qemu
@@ -1,4 +1,4 @@
-# Last Modified: Mon Apr  5 15:11:27 2010
+# Last Modified: Fri Mar 9 14:43:22 2012
 
   #include abstractions/base
   #include abstractions/consoles
@@ -108,3 +108,22 @@
   /bin/dash rmix,
   /bin/dd rmix,
   /bin/cat rmix,
+
+  /usr/libexec/qemu-bridge-helper Cx,
+  # child profile for bridge helper process
+  profile /usr/libexec/qemu-bridge-helper {
+   #include abstractions/base
+
+   capability setuid,
+   capability setgid,
+   capability setpcap,
+   capability net_admin,
+
+   network inet stream,
+
+   /dev/net/tun rw,
+   /etc/qemu/** r,
+   owner @{PROC}/*/status r,
+
+   /usr/libexec/qemu-bridge-helper rmix,
+  }
-- 
1.7.1

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [Patch v2 1/3] Add -netdev bridge capabilities

[rmarwah]

From: Richa Marwaha rmar...@linux.vnet.ibm.com

This patch adds the capability in libvirt to check if
-netdev bridge option is supported or not.

Signed-off-by: Richa Marwaha rmar...@linux.vnet.ibm.com
Signed-off-by: Corey Bryantcor...@linux.vnet.ibm.com
---
v2
-This is a new patch that helps libvirt to check if -netdev bridge
option is supported in the qemu version that is being used to run
the guest. In v1 we didnot have the capability to check if qemu
version we are using supported -netdev bridge or not.

 src/qemu/qemu_capabilities.c |   13 +
 src/qemu/qemu_capabilities.h |1 +
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
index 1e12a39..b2d7be2 100644
--- a/src/qemu/qemu_capabilities.c
+++ b/src/qemu/qemu_capabilities.c
@@ -166,6 +166,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST,
   hda-micro, /* 95 */
   dump-guest-memory,
   nec-usb-xhci,
+  bridge,
 
 );
 
@@ -1005,7 +1006,7 @@ qemuCapsComputeCmdFlags(const char *help,
 bool check_yajl ATTRIBUTE_UNUSED)
 {
 const char *p;
-const char *fsdev;
+const char *fsdev, *netdev;
 
 if (strstr(help, -no-kqemu))
 qemuCapsSet(flags, QEMU_CAPS_KQEMU);
@@ -1118,13 +1119,17 @@ qemuCapsComputeCmdFlags(const char *help,
 if (strstr(help, -smbios type))
 qemuCapsSet(flags, QEMU_CAPS_SMBIOS_TYPE);
 
-if (strstr(help, -netdev)) {
+if ((netdev = strstr(help, -netdev))) {
 /* Disable -netdev on 0.12 since although it exists,
  * the corresponding netdev_add/remove monitor commands
  * do not, and we need them to be able to do hotplug.
  * But see below about RHEL build. */
-if (version = 13000)
-qemuCapsSet(flags, QEMU_CAPS_NETDEV);
+if (version = 13000) {
+if (strstr (netdev, bridge))
+qemuCapsSet(flags, QEMU_CAPS_NETDEV_BRIDGE);
+else
+qemuCapsSet(flags, QEMU_CAPS_NETDEV);
+}
 }
 
 if (strstr(help, -sdl))
diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h
index 83c135b..458a73e 100644
--- a/src/qemu/qemu_capabilities.h
+++ b/src/qemu/qemu_capabilities.h
@@ -133,6 +133,7 @@ enum qemuCapsFlags {
 QEMU_CAPS_HDA_MICRO  = 95, /* -device hda-micro */
 QEMU_CAPS_DUMP_GUEST_MEMORY  = 96, /* dump-guest-memory command */
 QEMU_CAPS_NEC_USB_XHCI   = 97, /* -device nec-usb-xhci */
+QEMU_CAPS_NETDEV_BRIDGE  = 98, /* bridge helper support */
 
 QEMU_CAPS_LAST,   /* this must always be the last item */
 };
-- 
1.7.1

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [Patch v2 0/3] Add QEMU network helper support

[rmarwah]

From: Richa Marwaha rmar...@linux.vnet.ibm.com

QEMU has a new feature which allows QEMU to execute under an unprivileged user 
ID and still be able to
add a tap device to a Linux network bridge. Below is the link to the QEMU 
patches for the bridge helper
feature:

http://lists.gnu.org/archive/html/qemu-devel/2012-01/msg03562.html

The existing libvirt tap network device support for adding a tap device to a 
bridge (-netdev tap) works
only when connected to a libvirtd instance running as the privileged system 
account 'root'.
When connected to a libvirtd instance running as an unprivileged user (ie. 
using the session URI) creation of
the tap device fails as follows:

error: Failed to start domain F14_64 error: Unable to create tap device vnet%d: 
Operation not permitted

With this support, creating a tap device in the above scenario will be 
possible.  Additionally, hot attaching
a tap device to a bridge while running when connected to a libvirtd instance 
running as an unprivileged user
will be possible.

Richa Marwaha (3):
  Add -netdev bridge capabilities
  Add -netdev bridge support
  apparmor: QEMU bridge helper policy updates

 examples/apparmor/libvirt-qemu |   21 +-
 src/qemu/qemu_capabilities.c   |   13 ++--
 src/qemu/qemu_capabilities.h   |1 +
 src/qemu/qemu_command.c|   61 ---
 src/qemu/qemu_command.h|2 +
 src/qemu/qemu_hotplug.c|   31 ++--
 6 files changed, 97 insertions(+), 32 deletions(-)

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] nwfilter: Fix memory leak

[Stefan Berger]

On 06/28/2012 09:41 PM, Eric Blake wrote:

On 06/28/2012 06:38 PM, Stefan Berger wrote:

Below patch fixes this coverity report:

/libvirt/src/conf/nwfilter_conf.c:382:
leaked_storage: Variable varAccess going out of scope leaks the
storage it points to.

---
  src/conf/nwfilter_conf.c |1 +
  1 file changed, 1 insertion(+)

ACK.


pushed

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] availability of release candidate 2 of 0.9.13

[Ruben Kerkhof]

On Thu, Jun 28, 2012 at 5:36 AM, Daniel Veillard veill...@redhat.com wrote:
   I just tagged the git tree and I pushed the tarball for rc2:
     ftp://libvirt.org/libvirt/libvirt-0.9.13-rc2.tar.gz

Builds fine on OSX Lion.

Regards,

Ruben

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] availability of release candidate 2 of 0.9.13

[Jason Helfman]

On Fri, Jun 29, 2012 at 10:27:22PM +0200, Ruben Kerkhof thus spake:

On Thu, Jun 28, 2012 at 5:36 AM, Daniel Veillard veill...@redhat.com wrote:

  I just tagged the git tree and I pushed the tarball for rc2:
    ftp://libvirt.org/libvirt/libvirt-0.9.13-rc2.tar.gz


Builds fine on OSX Lion.

Regards,

Ruben


FreeBSD fails on this here:

include-dirs -Wunused -Wunknown-pragmas -Wstrict-aliasing -Wshadow 
-Wpointer-arith -Wbad-function-cast -Wcast-align -Wwrite-strings 
-Waggregate-return -Wstrict-prototypes -Wold-style-definition 
-Wmissing-prototypes -Wmissing-declarations -Wmissing-noreturn 
-Wmissing-format-attribute -Wredundant-decls -Wnested-externs -Winline 
-Winvalid-pch -Wvolatile-register-var -Wdisabled-optimization -Wattributes 
-Wdeprecated-declarations -Wdiv-by-zero -Wendif-labels -Wextra 
-Wformat-extra-args -Wformat-zero-length -Wformat=2 -Wmultichar 
-Wnormalized=nfc -Woverflow -Wpointer-to-int-cast -Wpragmas -Wtrampolines 
-Wno-missing-field-initializers -Wno-sign-compare -Wno-format-nonliteral 
-fexceptions -fasynchronous-unwind-tables -fdiagnostics-show-option 
-funit-at-a-time -fipa-pure-const -O2 -pipe -fno-strict-aliasing -MT 
libvirt_conf_la-domain_conf.lo -MD -MP -MF 
.deps/libvirt_conf_la-domain_conf.Tpo -c conf/domain_conf.c  -fPIC -DPIC -o 
.libs/libvirt_conf_la-domain_conf.o
conf/domain_conf.c: In function 'virDomainDiskDefParseXML':
conf/domain_conf.c:3662: warning: null format string [-Wformat]
conf/domain_conf.c:3679: warning: null format string [-Wformat]
conf/domain_conf.c: In function 'virDomainFSDefParseXML':
conf/domain_conf.c:4252: error: 'ULONG_LONG_MAX' undeclared (first use in this 
function)
conf/domain_conf.c:4252: error: (Each undeclared identifier is reported only 
once
conf/domain_conf.c:4252: error: for each function it appears in.)
conf/domain_conf.c:4314: warning: null format string [-Wformat]
conf/domain_conf.c:4320: warning: null format string [-Wformat]
conf/domain_conf.c: In function 'virDomainDefParseXML':
conf/domain_conf.c:7982: warning: null format string [-Wformat]
gmake[3]: *** [libvirt_conf_la-domain_conf.lo] Error 1
gmake[3]: Leaving directory 
`/usr/local/workspace/ports/devel/libvirt/work/libvirt-0.9.13/src'
gmake[2]: *** [all] Error 2
gmake[2]: Leaving directory 
`/usr/local/workspace/ports/devel/libvirt/work/libvirt-0.9.13/src'
gmake[1]: *** [all-recursive] Error 1
gmake[1]: Leaving directory 
`/usr/local/workspace/ports/devel/libvirt/work/libvirt-0.9.13'
gmake: *** [all] Error 2
*** Error code 1

-jgh

--
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] availability of release candidate 2 of 0.9.13

[Eric Blake]

On 06/29/2012 02:47 PM, Jason Helfman wrote:

 FreeBSD fails on this here:
 

 conf/domain_conf.c: In function 'virDomainFSDefParseXML':
 conf/domain_conf.c:4252: error: 'ULONG_LONG_MAX' undeclared (first use
 in this function)

POSIX requires limits.h to spell it ULLONG_MAX, not ULONG_LONG_MAX.
We've got a bug in our code for using the non-standard name; I'll
propose a patch soon.

-- 
Eric Blake   ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org





signature.asc
Description: OpenPGP digital signature
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] [PATCH] build: use correct limit for unsigned long long

[Eric Blake]

Reported by Jason Helfman as a build-breaker on FreeBSD.

* src/conf/domain_conf.c (virDomainFSDefParseXML): Use POSIX
spelling.
* src/openvz/openvz_conf.c (openvzReadFSConf): Likewise.
---

Pushing under the build-breaker rule.

 src/conf/domain_conf.c   |8 +++-
 src/openvz/openvz_conf.c |4 ++--
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index 4086dac..3fb90db 100644
--- a/src/conf/domain_conf.c
+++ b/src/conf/domain_conf.c
@@ -4248,14 +4248,12 @@ virDomainFSDefParseXML(xmlNodePtr node,

 if (virDomainParseScaledValue(./space_hard_limit[1], ctxt,
   def-space_hard_limit, 1,
-  ULONG_LONG_MAX,
-  false)  0)
+  ULLONG_MAX, false)  0)
 goto error;

 if (virDomainParseScaledValue(./space_soft_limit[1], ctxt,
   def-space_soft_limit, 1,
-  ULONG_LONG_MAX,
-  false)  0)
+  ULLONG_MAX, false)  0)
 goto error;

 cur = node-children;
@@ -4335,7 +4333,7 @@ virDomainFSDefParseXML(xmlNodePtr node,
 }
 if (unit 
 virScaleInteger(def-usage, unit,
-1024, ULONG_LONG_MAX)  0)
+1024, ULLONG_MAX)  0)
 goto error;
 }

diff --git a/src/openvz/openvz_conf.c b/src/openvz/openvz_conf.c
index bb61b13..ad27d37 100644
--- a/src/openvz/openvz_conf.c
+++ b/src/openvz/openvz_conf.c
@@ -455,8 +455,8 @@ openvzReadFSConf(virDomainDefPtr def,
 goto error;
 } else {
 /* Ensure that we can multiply by 1024 without overflowing. */
-if (barrier  ULONG_LONG_MAX / 1024 ||
-limit  ULONG_LONG_MAX / 1024 ) {
+if (barrier  ULLONG_MAX / 1024 ||
+limit  ULLONG_MAX / 1024 ) {
 virReportSystemError(VIR_ERR_OVERFLOW,
  _(%s),
  Unable to parse quota);
-- 
1.7.10.4

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] [PATCH] build: use correct limit for unsigned long long

[Jason Helfman]

On Fri, Jun 29, 2012 at 03:11:41PM -0600, Eric Blake thus spake:

Reported by Jason Helfman as a build-breaker on FreeBSD.


Patch is good. It builds, now.

Thanks!
-jgh

--
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] compile option: --without-gnutls

[Chris Van Heuveln]

Hi libvirt-list,

Full disclosure: this is the first open source project I've tried to contribute 
to and I'm still learning the process and tools.

In any case, Mikhail Gusarov started a thread on this a while back and I've 
tried to take it to conclusion. I needed to remove tls for 0.8.4 so I coded up 
a --without-gnutls configure option using Mikhail's partial code as a template. 
Then I pulled down 0.9.12 and came up with a patch that I'd like to push out to 
the commununity.

I'm not adding any new functionality other than the configure option. The tls 
calls/structs are just #ifdef'd for the most part. In a couple of places I had 
to flip the logic for if (!tls_struct) else, and also had to modify a few 
function calls to pass void * instead of tls pointers. 

I built --with-gnutls, --without-gnutls, and no option specified (defaults to 
--with-gnutls).

make check is clean except for:

TEST: libvirtdconftest
  .!!...!  39  FAIL

...which consists mostly of this sasl failure:

39) Test corruption   ... 
libvir: Config File error : unsupported configuration: remoteReadConfigFile: 
/home/ubu/git/libvirt/tests/../daemon/libvirtd.conf: auth_tcp: unsupported auth 
sasl
FAILED

...so I'm trying to track that down, and also need to run syntax-check and 
valgrind tests.

Once I get the tests passing what's my next step? 
Are you okay with this approach? 
Do you want a preview of my diffs? 
Do I need to write additional tests for this new option or update any 
documentation? 

Thanks,
Chris Van Heuveln






--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

[libvirt] sanlock on F17

[Dave Allan]

I just tried to set up sanlock on F17 using the instructions at
http://libvirt.org/locking.html, but libvirtd refuses to start with
the sanlock error in the logs:

Jun 29 19:56:20 nienna sanlock[8423]: 19846 open error -13
/var/lib/libvirt/sanlock/__LIBVIRT__DISKS__

I confirmed that user sanlock can create, read and write files in
/var/lib/libvirt/sanlock

The only thing I can think of that's slightly odd about my setup is
that I'm only concerned with the local machine, so that directory is
not an NFS mount.  Anybody have any thoughts on what's going wrong
here?

TIA,
Dave

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] availability of release candidate 2 of 0.9.13

[Jason Helfman]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On Fri, Jun 29, 2012 at 03:07:52PM -0600, Eric Blake thus spake:
On 06/29/2012 02:47 PM, Jason Helfman wrote:

 FreeBSD fails on this here:


 conf/domain_conf.c: In function 'virDomainFSDefParseXML':
 conf/domain_conf.c:4252: error: 'ULONG_LONG_MAX' undeclared (first use
 in this function)

POSIX requires limits.h to spell it ULLONG_MAX, not ULONG_LONG_MAX.
We've got a bug in our code for using the non-standard name; I'll
propose a patch soon.


Here is a link to the build logs with the patch:
https://redports.org/buildarchive/20120630004335-48297/

- -jgh

- -- 
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (FreeBSD)

iF4EAREIAAYFAk/uTW0ACgkQXpKtZoyM+6X4swD+Ib387L0WgF0VumhmWRxBHb6Y
xXr3aYw6bheKGv0BopcA/jJjNhnBgXzeLUvbNmthcOt5n+uTTnXDKYQpd3eEDMvQ
=sIXU
-END PGP SIGNATURE-

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] availability of release candidate 2 of 0.9.13

[Doug Goldstein]

On Wed, Jun 27, 2012 at 10:36 PM, Daniel Veillard veill...@redhat.com wrote:
   I just tagged the git tree and I pushed the tarball for rc2:
     ftp://libvirt.org/libvirt/libvirt-0.9.13-rc2.tar.gz

  the rpms are being built and will be pushed soon too
 I didn't get any feedback on other platforms or OSes for rc1,
 hopefully this will get a bit of testing before the actual
 release, probably on Monday early next week.

  thanks !

 Daniel


Build fine on Gentoo, however it fails a test. Here's the stdout from
the overall run, I'll dig in deeper over the weekend.

TEST: virdrivermoduletest
  !!.  7   FAIL

-- 
Doug Goldstein

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

Re: [libvirt] sanlock on F17

[Dave Allan]

On Fri, Jun 29, 2012 at 08:06:34PM -0400, Dave Allan wrote:
 I just tried to set up sanlock on F17 using the instructions at
 http://libvirt.org/locking.html, but libvirtd refuses to start with
 the sanlock error in the logs:
 
 Jun 29 19:56:20 nienna sanlock[8423]: 19846 open error -13
 /var/lib/libvirt/sanlock/__LIBVIRT__DISKS__
 
 I confirmed that user sanlock can create, read and write files in
 /var/lib/libvirt/sanlock

Also disabling selinux does not help.

 The only thing I can think of that's slightly odd about my setup is
 that I'm only concerned with the local machine, so that directory is
 not an NFS mount.  Anybody have any thoughts on what's going wrong
 here?
 
 TIA,
 Dave

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list