Roger Pau Monné | 10 Feb 16:22
Favicon

qemu-xen qdisk performance

Hello,

I've recently setup a Linux Dom0 with a 3.0.17 kernel and Xen 4.1.2,
and since the 3.x series doesn't have blktap support I'm using qdisk
to attach raw images. I've been playing with small images, something
like 1GB, and everything seemed fine, speed was not fantastic but it
was ok. Today I've set up a bigger machine, with a 20GB raw hdd and
the disk write throughput is really slow, inferior to 0.5MB/s. I'm
trying to install a Debian PV there, and after more than 3 hours it is
still installing the base system.

I've looked at the xenstore backend entries, and everything looks fine:

/local/domain/0/backend/qdisk/21/51712/frontend =
"/local/domain/21/device/vbd/51712"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/params =
"aio:/hdd/vm/servlet/servlet.img"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/frontend-id = "21"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/online = "1"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/removable = "0"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/bootable = "1"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/state = "4"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/dev = "xvda"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/type = "tap"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/mode = "w"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/feature-barrier = "1"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/info = "0"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/sector-size = "512"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/sectors = "40960000"   (n0,r21)
/local/domain/0/backend/qdisk/21/51712/hotplug-status = "connected"   (n0,r21)
(Continue reading)

Wei Wang | 10 Feb 16:07
Picon

[PATCH 0 of 6 V5] amd iommu: support ats/gpgpu passthru on iommuv2 systems

Hi,
This is patch set v5. It includes all pending patches that are needed to enable gpgpu passthrough and
heterogeneous computing in guest OSes. Basically, this patch set gives guest VM the same capability of
running openCL applications on amd platforms as native OSes. Upstream Linux 3.3 rc2 with amd iommuv2
kernel driver has been tested well as guest OS, and since last submission, lots of regression tests have
been done to make sure this does not break non-iommuv2 systems. Please review it, feedbacks are appreciated.

Many thanks,
Wei

For more details, please refer to old thread.
http://lists.xen.org/archives/html/xen-devel/2012-01/msg01646.html

and, for an overview of the design, please refer to
http://www.amd64.org/pub/iommuv2.png

======================================================================
changes in v5:
* Remove patch 2 after upstream c/s 24729:6f6a6d1d2fb6

changes in v4:
* Only tool part in this version, since hypervisor patches have already been committed.
* rename guest config option from "iommu = {0,1}" to "guest_iommu = {0,1}"
* add description into docs/man/xl.cfg.pod.5

changes in v3:
* Use xenstore to receive guest iommu configuration instead of adding in a new field in hvm_info_table.
* Support pci segment in vbdf to mbdf bind.
* Make hypercalls visible for non-x86 platforms.
* A few code cleanups according to comments from Jan and Ian.
(Continue reading)

young zhang | 10 Feb 14:39
Picon
Gravatar

how to add a new trace event?

I want to trace something which is not covered by current events,  but
i found the logic is very complicated. Is there any simple example of
adding a new trace event?

--

-- 
young
xen.org | 10 Feb 14:32

[xen-unstable bisection] complete test-amd64-amd64-win

branch xen-unstable
xen branch xen-unstable
job test-amd64-amd64-win
test windows-install

Tree: linux git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen.git
Tree: qemu git://xenbits.xen.org/staging/qemu-xen-unstable.git
Tree: qemuu git://xenbits.xen.org/staging/qemu-upstream-unstable.git
Tree: xen http://xenbits.xen.org/staging/xen-unstable.hg

*** Found and reproduced problem changeset ***

  Bug is in tree:  xen http://xenbits.xen.org/staging/xen-unstable.hg
  Bug introduced:  aae516b78fce
  Bug not present: 585caf50a926

  changeset:   24757:aae516b78fce
  user:        Alex Zeffertt <alex.zeffertt <at> eu.citrix.com>
  date:        Thu Feb 09 18:33:32 2012 +0000

      xenstored: use grant references instead of map_foreign_range

      make xenstored use grantref rather than map_foreign_range (which can
      only be used by privileged domains)

      This patch modifies the xenstore daemon to use xc_gnttab_map_grant_ref
      instead of xc_map_foreign_range where available.

      Previous versions of this patch have been sent to xen-devel. See
      http://lists.xensource.com/archives/html/xen-devel/2008-07/msg00610.html
(Continue reading)

Ian Campbell | 10 Feb 14:04
Favicon

[PATCH] xl: Add -F to usage for xl shutdown/reboot

# HG changeset patch
# User Ian Campbell <ian.campbell <at> citrix.com>
# Date 1328879048 0
# Node ID cc609327d4fba381388be7826760017f02d1748f
# Parent  13621819e3c7b3a80cc3cf5a092585fae21751d7
xl: Add -F to usage for xl shutdown/reboot

Signed-off-by: Ian Campbell <ian.campbell <at> citrix.com>

diff -r 13621819e3c7 -r cc609327d4fb tools/libxl/xl_cmdtable.c
--- a/tools/libxl/xl_cmdtable.c	Fri Feb 10 12:08:18 2012 +0000
+++ b/tools/libxl/xl_cmdtable.c	Fri Feb 10 13:04:08 2012 +0000
@@ -48,12 +48,19 @@ struct cmd_spec cmd_table[] = {
     { "shutdown",
       &main_shutdown, 0,
       "Issue a shutdown signal to a domain",
-      "<Domain>",
+      "[options] <Domain>",
+      "-h                      Print this help.\n"
+      "-F                      Fallback to ACPI power event for HVM guests with\n"
+      "                        no PV drivers.\n"
+      "-w                      Wait for guest to shutdown.\n"
     },
     { "reboot",
       &main_reboot, 0,
       "Issue a reboot signal to a domain",
-      "<Domain>",
+      "[options] <Domain>",
+      "-h                      Print this help.\n"
+      "-F                      Fallback to ACPI reset event for HVM guests with\n"
(Continue reading)

David Vrabel | 10 Feb 14:03
Favicon

[PATCH 0/8] arm: initial device tree support (#2)

This series adds preliminary device tree support to Xen for ARM.

libfdt for parsing a device tree in flattened device tree (fdt) format
and added and used to find the location and size of RAM.  This info is
then used when relocating Xen and setting up the mm and heaps.

Since we don't have a bootloader when using the model the device tree
blob (DTB) is linked into .dtb section of the Xen image.  The DTB
needs to be present in xen/arch/arm/platfom.dtb to build.  See the
wiki[1] for the Linux kernel which has a DTB for the model.

This series is also available in the device-tree-v4 branch of
git://xenbits.xen.org/people/dvrabel/xen-unstable.git

Changes since v1:

- map DTB in head.S using the L2 slot after the fixmap.
- new patch 8 for setting up mm/heaps.

David

[1] http://wiki.xen.org/wiki/Xen_ARMv7_with_Virtualization_Extensions 
xen.org | 10 Feb 14:02

[xen-unstable test] 11918: regressions - FAIL

flight 11918 xen-unstable real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/11918/

Regressions :-(

Tests which did not succeed and are blocking,
including tests which could not be run:
 test-amd64-amd64-win          7 windows-install  fail in 11911 REGR. vs. 11904
 test-amd64-i386-win-vcpus1    7 windows-install  fail in 11911 REGR. vs. 11904
 test-amd64-i386-win           7 windows-install  fail in 11911 REGR. vs. 11904
 test-amd64-i386-xend-winxpsp3  7 windows-install fail in 11911 REGR. vs. 11904
 test-i386-i386-win            7 windows-install  fail in 11911 REGR. vs. 11904

Tests which are failing intermittently (not blocking):
 build-amd64-oldkern           4 xen-build                   fail pass in 11911
 build-i386                    4 xen-build                   fail pass in 11911
 build-i386-oldkern            4 xen-build                   fail pass in 11911
 build-amd64                   4 xen-build                   fail pass in 11911
 test-amd64-amd64-xl-sedf-pin 14 guest-localmigrate/x10 fail in 11911 pass in 11907

Tests which did not succeed, but are not blocking:
 test-i386-i386-xl-qemuu-winxpsp3  1 xen-build-check(1)           blocked  n/a
 test-amd64-amd64-xl-qemuu-winxpsp3  1 xen-build-check(1)           blocked n/a
 test-amd64-amd64-xl-qemuu-win7-amd64  1 xen-build-check(1)         blocked n/a
 test-amd64-i386-qemuu-rhel6hvm-intel  1 xen-build-check(1)         blocked n/a
 test-amd64-amd64-xl-pcipt-intel  1 xen-build-check(1)           blocked  n/a
 test-amd64-i386-qemuu-rhel6hvm-amd  1 xen-build-check(1)           blocked n/a
 test-amd64-amd64-xl           1 xen-build-check(1)           blocked  n/a
 test-amd64-amd64-pv           1 xen-build-check(1)           blocked  n/a
 test-amd64-amd64-xl-sedf-pin  1 xen-build-check(1)           blocked  n/a
(Continue reading)

xen.org | 10 Feb 13:12

[xen-unstable test] 11911: regressions - FAIL

flight 11911 xen-unstable real [real]
http://www.chiark.greenend.org.uk/~xensrcts/logs/11911/

Regressions :-(

Tests which did not succeed and are blocking,
including tests which could not be run:
 test-amd64-amd64-win          7 windows-install           fail REGR. vs. 11904
 test-amd64-i386-win-vcpus1    7 windows-install           fail REGR. vs. 11904
 test-amd64-i386-win           7 windows-install           fail REGR. vs. 11904
 test-amd64-i386-xend-winxpsp3  7 windows-install          fail REGR. vs. 11904
 test-i386-i386-win            7 windows-install           fail REGR. vs. 11904

Tests which are failing intermittently (not blocking):
 test-amd64-amd64-xl-sedf-pin 14 guest-localmigrate/x10      fail pass in 11907

Tests which did not succeed, but are not blocking:
 test-i386-i386-xl-qemuu-winxpsp3  7 windows-install            fail never pass
 test-amd64-amd64-xl-qemuu-winxpsp3  7 windows-install          fail never pass
 test-amd64-amd64-xl-qemuu-win7-amd64  7 windows-install        fail never pass
 test-amd64-i386-qemuu-rhel6hvm-intel  7 redhat-install         fail never pass
 test-amd64-amd64-xl-pcipt-intel  9 guest-start                 fail never pass
 test-amd64-i386-qemuu-rhel6hvm-amd  7 redhat-install           fail never pass
 test-amd64-amd64-xl-winxpsp3 13 guest-stop                   fail   never pass
 test-amd64-i386-rhel6hvm-intel 11 leak-check/check             fail never pass
 test-amd64-i386-rhel6hvm-amd 11 leak-check/check             fail   never pass
 test-i386-i386-xl-winxpsp3   13 guest-stop                   fail   never pass
 test-amd64-amd64-xl-win7-amd64 13 guest-stop                   fail never pass
 test-amd64-i386-xl-winxpsp3-vcpus1 13 guest-stop               fail never pass
 test-i386-i386-xl-win        13 guest-stop                   fail   never pass
(Continue reading)

Stefano Stabellini | 10 Feb 13:01
Favicon

[PATCH v2 00/10] arm: compile tools

Hi all,
this patch series allows tools/ to compile on ARM, mostly providing an
empty implementation for all the arch specific functions that are needed.

Changes in v2:

- rebased on a22587ae517170a7755d3a88611ae0e2d5bb555e;

- dropped "arm: arch_dump_shared_mem_info as a no-op" that is already in
xen-unstable;

- define xen_callback_t as uint64_t;

- define guest_word_t as uint64_t.

Ian Campbell (4):
      arm: add stub hvm/save.h
      libxl: do not allocate e820 for non x86 guests.
      blktap2/libvhd: Build shared objects using -fPIC.
      tools: only compile libfsimage/xfs on X86

Stefano Stabellini (6):
      arm: few missing #define
      arm: compile libxc
      arm: compile libxenguest
      arm: compile libxl
      arm: compile memshr
      arm: compile xentrace
      arm: compile xentrace

(Continue reading)

George Dunlap | 10 Feb 12:06
Favicon

[PATCH] qemu: Don't access /proc/bus/pci unless graphics pass-thru is enabled

A recent changeset introduced a bug whereby an initialization function
that reads /proc/bus/pci is called from graphics set-up functions
even if pass-through graphics are not enabled.  If qemu is run without
permission to this file, this causes qemu to fail during initialization.

This patch changes the initialization functions to only happen if graphics
pass-through are enabled.

Signed-off-by: George Dunlap <george.dunlap <at> eu.citrix.com>
Attachment (fix-intel_pch_init.diff): text/x-diff, 2094 bytes
_______________________________________________
Xen-devel mailing list
Xen-devel <at> lists.xensource.com
http://lists.xensource.com/xen-devel
Frank, Chen | 6 Feb 02:00
Favicon
Gravatar

Re: [XenARM] [help] How to compile the unstable source for arm at sstabellini/xen-unstable.git/.git

Hi Xen-ARMs,
 
I applied this patch, "error: unknown type name 'xen_callback_t' is fixed. But I still meet the problem. The problem is where to download the cross compiler "arm-none-linux-gnueabi-gcc" which supports march=arm-V7a and mcpu=cortex-a15.
 
Would someone give me one avaiable URL to download the right compiler? 
Best Regards,
Frank
--


At 2012-01-31 22:51:29,"Stefano Stabellini" <stefano.stabellini <at> eu.citrix.com> wrote: >On Tue, 31 Jan 2012, Ian Campbell wrote: >> > make[3]: Entering directory >> > `/home/frank/workspace/xen/src/xen-arm-v6/xen/include' >> > for i in public/callback.h public/dom0_ops.h public/elfnote.h >> > public/event_channel.h public/features.h public/grant_table.h >> > public/kexec.h public/mem_event.h public/memory.h public/nmi.h >> > public/physdev.h public/platform.h public/sched.h public/tmem.h >> > public/trace.h public/vcpu.h public/version.h public/xen-compat.h >> > public/xen.h public/xencomm.h public/xenoprof.h public/hvm/e820.h >> > public/hvm/hvm_info_table.h public/hvm/hvm_op.h public/hvm/ioreq.h >> > public/hvm/params.h public/io/blkif.h public/io/console.h >> > public/io/fbif.h public/io/fsif.h public/io/kbdif.h >> > public/io/libxenvchan.h public/io/netif.h public/io/pciif.h >> > public/io/protocols.h public/io/ring.h public/io/tpmif.h >> > public/io/usbif.h public/io/vscsiif.h public/io/xenbus.h >> > public/io/xs_wire.h; do arm-linux-gnueabi-gcc -ansi -include stdint.h >> > -Wall -W -Werror -S -o /dev/null -xc $i || exit 1; echo $i; done >> > >headers.chk.new >> > public/callback.h:87:5: error: unknown type name 'xen_callback_t' >>  >> At this point I get: >> [ -e include/asm ] || ln -sf asm-arm include/asm >> make -f /local/scratch/ianc/devel/arm/xen-unstable/xen/Rules.mk -C include >> make[3]: Entering directory `/local/scratch/ianc/devel/arm/xen-unstable/xen/include' >> make[3]: Nothing to be done for `all'. >> make[3]: Leaving directory `/local/scratch/ianc/devel/arm/xen-unstable/xen/include' >>  >> Aha -- the difference is down to XEN_TARGET_ARCH vs. XEN_COMPILE_ARCH, >> see towards the end of xen/include/Makefile: >>         ifeq ($(XEN_TARGET_ARCH),$(XEN_COMPILE_ARCH)) >>         ... >>         all: headers.chk >>         ... >>  >> I'm inferring from the Makefile that: >> XEN_COMPILE_ARCH == the host architecture -- e.g. the machine you are >>  compiling on >> XEN_TARGET_ARCH == the target architecture -- e.g. the machine you want  >>  to run the resulting Xen on. >>  >> And indeed if I do a native build on an arm system I see the same error >> as you do. We'll look at fixing this but in the meantime I suggest you >> use XEN_TARGET_ARCH and not XEN_COMPILE_ARCH. > >The following patch fixes the compile issue (that indeed is due to the >header files check you pointed out). > >--- > >arm: few missing #define > >Few missing #define are the cause of a compile failure with >XEN_TARGET_ARM=arm and XEN_COMPILE_ARM=arm (for example in the case of a >native compilation). This patch fill the gaps. > >Signed-off-by: Stefano Stabellini <stefano.stabellini <at> eu.citrix.com> > >diff --git a/xen/include/public/arch-arm.h b/xen/include/public/arch-arm.h >index c430cf3..e04c4fd 100644 >--- a/xen/include/public/arch-arm.h >+++ b/xen/include/public/arch-arm.h > <at> <at>  -110,6 +110,8  <at> <at>  typedef struct arch_vcpu_info arch_vcpu_info_t; >  > struct arch_shared_info { }; > typedef struct arch_shared_info arch_shared_info_t; >+typedef unsigned long xen_callback_t; >+ > #endif >  > #endif /*  __XEN_PUBLIC_ARCH_ARM_H__ */ >diff --git a/xen/include/public/io/protocols.h b/xen/include/public/io/protocols.h >index 77bd1bd..0b7a2ea 100644 >--- a/xen/include/public/io/protocols.h >+++ b/xen/include/public/io/protocols.h > <at> <at>  -26,6 +26,7  <at> <at> > #define XEN_IO_PROTO_ABI_X86_32     "x86_32-abi" > #define XEN_IO_PROTO_ABI_X86_64     "x86_64-abi" > #define XEN_IO_PROTO_ABI_IA64       "ia64-abi" >+#define XEN_IO_PROTO_ABI_ARM        "arm-abi" >  > #if defined(__i386__) > # define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_X86_32 > <at> <at>  -33,6 +34,8  <at> <at> > # define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_X86_64 > #elif defined(__ia64__) > # define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_IA64 >+#elif defined(__arm__) >+# define XEN_IO_PROTO_ABI_NATIVE XEN_IO_PROTO_ABI_ARM > #else > # error arch fixup needed here > #endif


_______________________________________________
Xen-devel mailing list
Xen-devel <at> lists.xensource.com
http://lists.xensource.com/xen-devel

Gmane