xen.git
19 years ago[IOEMU] Locking needed to access mapcache.
kfraser@localhost.localdomain [Wed, 13 Dec 2006 11:01:33 +0000 (11:01 +0000)]
[IOEMU] Locking needed to access mapcache.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Use cpumask macros to update numa node masks.
kfraser@localhost.localdomain [Wed, 13 Dec 2006 10:24:20 +0000 (10:24 +0000)]
[XEN] Use cpumask macros to update numa node masks.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[TOOLS] Fix the build. Clearly demarcate PPC-specific stuff.
kfraser@localhost.localdomain [Wed, 13 Dec 2006 10:23:53 +0000 (10:23 +0000)]
[TOOLS] Fix the build. Clearly demarcate PPC-specific stuff.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoTiny fix.
Steven Hand [Wed, 13 Dec 2006 10:14:40 +0000 (10:14 +0000)]
Tiny fix.

Signed-off-by: Steven Hand <steven@xensource.com>
19 years agoUpdate ignores.
Steven Hand [Wed, 13 Dec 2006 10:14:17 +0000 (10:14 +0000)]
Update ignores.

Signed-off-by: Steven Hand <steven@xensource.com>
19 years agoMerge with xenppc-unstable
kfraser@localhost.localdomain [Wed, 13 Dec 2006 10:05:07 +0000 (10:05 +0000)]
Merge with xenppc-unstable

19 years ago[XEN] Fix kexec crash.
kfraser@localhost.localdomain [Wed, 13 Dec 2006 09:46:48 +0000 (09:46 +0000)]
[XEN] Fix kexec crash.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[POWERPC][XEN] Add support for || as a xen/dom0 commandline divider.
Hollis Blanchard [Tue, 12 Dec 2006 20:35:07 +0000 (14:35 -0600)]
[POWERPC][XEN] Add support for || as a xen/dom0 commandline divider.
Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
--HG--
extra : transplant_source : %DB%C7M%B1JK9%D3Y6_%CF%82W%21m%96%8F%A2i

19 years ago[XEN] Improve multi-core/thread scheduler support.
Emmanuel Ackaouy [Tue, 12 Dec 2006 19:37:21 +0000 (19:37 +0000)]
[XEN] Improve multi-core/thread scheduler support.
Relax granularity at which running VCPUs are migrated onto
cores or sockets with more idle execution vehicles and
simplify code a good deal in the process.
Also, some scheduler stat cleanup while in there.
Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
19 years ago[XEN] Be sure to discard old PAE l2 shadows when guest l2e becomes null.
Tim Deegan [Tue, 12 Dec 2006 16:52:13 +0000 (16:52 +0000)]
[XEN] Be sure to discard old PAE l2 shadows when guest l2e becomes null.
This fixes recent PAE live migrate failures.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[XEN][POWERPC] boot_of_allocator fixups
Jimi Xenidis [Tue, 12 Dec 2006 01:50:32 +0000 (20:50 -0500)]
[XEN][POWERPC] boot_of_allocator fixups
This patch:
  - print devtree mod size correctly
  - find next_bit rather then test all needed bits to be zero
  - stop pre-allocating memory below image as used
Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
--HG--
extra : transplant_source : %88%7E%1C%BA%C6%15M%A0%A3%A3%C2C%3F%BC%5B%0F%C2%A1%C9%B8

19 years ago[XEN][POWERPC] Fix systemsim-gpul failure to boot
Jimi Xenidis [Tue, 12 Dec 2006 00:43:18 +0000 (19:43 -0500)]
[XEN][POWERPC] Fix systemsim-gpul failure to boot
This patch fixes the fencepost error and adds a check for the failure to
initialize the allocator, which would have make Mark's FTB much easier
to debug.
This works for me on the public systemsim-gpul release.
Signed-off-by: Amos Waterland <apw@us.ibm.com>
Also fixes Maple, was just getting luck.
Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
--HG--
extra : transplant_source : %87%8C%E1%F7%8A%D3%97oD%A6W%E58j%7Fma%9F%FA%C2

19 years ago[POWERPC][XEN] Remove unused uaccess.h header.
Hollis Blanchard [Tue, 12 Dec 2006 19:54:00 +0000 (13:54 -0600)]
[POWERPC][XEN] Remove unused uaccess.h header.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Fix gdbstub build break.
Hollis Blanchard [Tue, 12 Dec 2006 19:52:38 +0000 (13:52 -0600)]
[POWERPC][XEN] Fix gdbstub build break.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Fix kexec build breakage.
Hollis Blanchard [Tue, 12 Dec 2006 19:52:18 +0000 (13:52 -0600)]
[POWERPC][XEN] Fix kexec build breakage.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Fix build break by using bitmap field in cpumask_t.
Hollis Blanchard [Tue, 12 Dec 2006 19:51:29 +0000 (13:51 -0600)]
[POWERPC][XEN] Fix build break by using bitmap field in cpumask_t.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Fix warning by removing "packed".
Hollis Blanchard [Tue, 12 Dec 2006 17:17:27 +0000 (11:17 -0600)]
[POWERPC][XEN] Fix warning by removing "packed".
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Build fixes.
Hollis Blanchard [Tue, 12 Dec 2006 16:23:58 +0000 (10:23 -0600)]
[POWERPC][XEN] Build fixes.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Add empty xenoprof.h.
Hollis Blanchard [Tue, 12 Dec 2006 16:23:25 +0000 (10:23 -0600)]
[POWERPC][XEN] Add empty xenoprof.h.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC] Merge with xen-unstable.hg.
Hollis Blanchard [Tue, 12 Dec 2006 15:58:42 +0000 (09:58 -0600)]
[POWERPC] Merge with xen-unstable.hg.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC] Merge in outstanding changes from xenppc-unstable.hg.
Hollis Blanchard [Tue, 12 Dec 2006 15:56:06 +0000 (09:56 -0600)]
[POWERPC] Merge in outstanding changes from xenppc-unstable.hg.
It's a long story, but basically a small divergence in xenppc-unstable meant a
large number of changesets couldn't be directly imported to xen-unstable, so
this changeset includes all of them.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years agoMerge.
ssmith@localhost.localdomain [Mon, 11 Dec 2006 19:17:11 +0000 (11:17 -0800)]
Merge.

19 years ago[XEN] The shadow FAST_FAULT_PATH optimisation assumes that pages never
ssmith@localhost.localdomain [Mon, 11 Dec 2006 19:16:29 +0000 (11:16 -0800)]
[XEN] The shadow FAST_FAULT_PATH optimisation assumes that pages never
transition between mmio and RAM-backed.  This isn't true after a
an add_to_physmap memory op.  Fix this by just blowing the shadow tables
after every such operation; they're rare enough that the performance
hit is not a concern.

Signed-off-by: Steven Smith <sos22@cam.ac.uk>
Acked-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[XEN] Only allow each domain to putput crash debug info once.
kfraser@localhost.localdomain [Mon, 11 Dec 2006 16:32:25 +0000 (16:32 +0000)]
[XEN] Only allow each domain to putput crash debug info once.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Remove unnecessary shadow_mode_refcount() code from PV mm code.
kfraser@localhost.localdomain [Mon, 11 Dec 2006 15:06:53 +0000 (15:06 +0000)]
[XEN] Remove unnecessary shadow_mode_refcount() code from PV mm code.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[BUILDER] Enable shadow mode for shadow translate guests much earlier.
Ian Campbell [Mon, 11 Dec 2006 11:57:36 +0000 (11:57 +0000)]
[BUILDER] Enable shadow mode for shadow translate guests much earlier.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[NET] back: Atomic buffer allocations use __GFP_NOWARN.
kfraser@localhost.localdomain [Mon, 11 Dec 2006 09:56:02 +0000 (09:56 +0000)]
[NET] back: Atomic buffer allocations use __GFP_NOWARN.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Shadow-mode-refcount PTE update fix.
kaf24@localhost.localdomain [Sat, 9 Dec 2006 16:29:52 +0000 (16:29 +0000)]
[XEN] Shadow-mode-refcount PTE update fix.

Add back in support for emulated PTE updates which is critical for
shdow_refcount operation.

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agopygrub tmp files should live in /var/run/ not /var/lib/, as they are
kaf24@localhost.localdomain [Sat, 9 Dec 2006 15:04:27 +0000 (15:04 +0000)]
pygrub tmp files should live in /var/run/ not /var/lib/, as they are
indeed runtime only.

Also fix a race condition in making the pygrub fifo.

Signed-off-by: John Levon <john.levon@sun.com>
19 years agoMerge with xen-ia64-unstable.hg
kaf24@localhost.localdomain [Sat, 9 Dec 2006 14:34:53 +0000 (14:34 +0000)]
Merge with xen-ia64-unstable.hg

19 years ago[qemu patches] Update patches upto changeset 12814:c3ad3fcfe3.
Christian Limpach [Fri, 8 Dec 2006 18:37:19 +0000 (18:37 +0000)]
[qemu patches] Update patches upto changeset 12814:c3ad3fcfe3.

Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
19 years ago[IA64] avoid freeing arch.privregs twice
awilliam@xenbuild2.aw [Fri, 8 Dec 2006 18:31:29 +0000 (11:31 -0700)]
[IA64] avoid freeing arch.privregs twice

Not sure how this worked before, but mapping vcpu regs into
guest pseudo phys space causes extra problems.

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years ago[qemu patches] Update patches upto changeset 12756:1d32fb45e0.
Christian Limpach [Fri, 8 Dec 2006 18:31:01 +0000 (18:31 +0000)]
[qemu patches] Update patches upto changeset 12756:1d32fb45e0.

Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
19 years ago[IA64] change XEN_DOMCTL_getmemlist to return gmfn instead of mfn
awilliam@xenbuild2.aw [Fri, 8 Dec 2006 16:32:15 +0000 (09:32 -0700)]
[IA64] change XEN_DOMCTL_getmemlist to return gmfn instead of mfn

This patch tries to get 'xm dump-core' work on ia64 again.
With this patch, 'xm dump-core' results in creating core-dump file
as before with a slight format changes.
(machine address array -> pseudo physicall address array)
However xm dump-core has been broken on ia64 since
XEN_DOMCTL_getmemlist/ia64 is differenct from x86's.
This patch does NOT fix it.

Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years agoxm block-attach can take an optional 5th argument, show it in the help.
john.levon@sun.com [Fri, 8 Dec 2006 15:22:21 +0000 (07:22 -0800)]
xm block-attach can take an optional 5th argument, show it in the help.

Signed-off-by: John Levon <john.levon@sun.com>
19 years ago[LIBXC] Make strerror() thread-safe by protecting it with a mutex.
kfraser@localhost.localdomain [Fri, 8 Dec 2006 15:59:30 +0000 (15:59 +0000)]
[LIBXC] Make strerror() thread-safe by protecting it with a mutex.
Using strerror_r() is made difficult by the different GNU definition,
and different distros variously choose the POSIX or GNU prototype.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Keep vif IP addresses as space separated string.
Alastair Tse [Fri, 8 Dec 2006 13:31:21 +0000 (13:31 +0000)]
[XEND] Keep vif IP addresses as space separated string.

They were inconsistently represented as strings or lists making it
difficult to know the type of the IP parameter for vifs. Since both xm
and DevController represents these as space separated strings, that is
what it will be stored as.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[XEND] Make sure UUID is in the right format.
Alastair Tse [Fri, 8 Dec 2006 13:28:22 +0000 (13:28 +0000)]
[XEND] Make sure UUID is in the right format.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[LIBXC] Clean up use of sterror(). Define a thread-safe
kfraser@localhost.localdomain [Fri, 8 Dec 2006 11:57:06 +0000 (11:57 +0000)]
[LIBXC] Clean up use of sterror(). Define a thread-safe
version which uses strerror_r(), and tries to cope with
the POSIX and GNU versions of that function.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[LINUX] Kexec: Remove remainder of kexec-generic.patch
Ian Campbell [Fri, 8 Dec 2006 11:47:09 +0000 (11:47 +0000)]
[LINUX] Kexec: Remove remainder of kexec-generic.patch

Rename xen_machine_kexec() to just machine_kexec() and only compile in
the Xen or native version as appropriate.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[LINUX] Kexec: Reapply kexec patches to sparse tree.
Ian Campbell [Fri, 8 Dec 2006 11:47:09 +0000 (11:47 +0000)]
[LINUX] Kexec: Reapply kexec patches to sparse tree.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[LINUX] Kexec: add kexec files to sparse tree.
Ian Campbell [Fri, 8 Dec 2006 11:47:09 +0000 (11:47 +0000)]
[LINUX] Kexec: add kexec files to sparse tree.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[LINUX] kexec: remove patches which are not straight backports.
Ian Campbell [Fri, 8 Dec 2006 11:47:09 +0000 (11:47 +0000)]
[LINUX] kexec: remove patches which are not straight backports.

Their contents will be added to the sparse tree in a subsequent patch.

Changes to kernel/sys.c in kexec-generic.patch remain for now. These
will be taken care of in a later patch.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[LINUX] Kexec: Do not bug if a kexec hypercall fails.
Ian Campbell [Fri, 8 Dec 2006 11:46:48 +0000 (11:46 +0000)]
[LINUX] Kexec: Do not bug if a kexec hypercall fails.

This is non fatal since we can recover and boot without kexec
functionality.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[XEN] Implement XENMEM_set_memory_map, which specifies memory map to
kfraser@localhost.localdomain [Fri, 8 Dec 2006 11:30:30 +0000 (11:30 +0000)]
[XEN] Implement XENMEM_set_memory_map, which specifies memory map to
be returned by XENMEM_memory_map. Hook this into the domain builder.

Based on a patch by Glauber de Oliveira Costa <gcosta@redhat.com>

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Fix xm vcpu-set when decrementing VCPU count
Alastair Tse [Fri, 8 Dec 2006 11:08:26 +0000 (11:08 +0000)]
[XEND] Fix xm vcpu-set when decrementing VCPU count

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years agoFix potential minor memory leak.
Steven Hand [Fri, 8 Dec 2006 10:48:46 +0000 (10:48 +0000)]
Fix potential minor memory leak.

Signed-off-by: Steven Hand <steven@xensource.com>
19 years agoDownload and symlink the i386 version of the ramdisk when running on x86_64.
Ewan Mellor [Fri, 8 Dec 2006 10:32:48 +0000 (10:32 +0000)]
Download and symlink the i386 version of the ramdisk when running on x86_64.
This means that people don't need to explicitly set BR_ARCH.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years ago[LINUX] CONFIG_XEN_BACKEND=m build properly.
kfraser@localhost.localdomain [Fri, 8 Dec 2006 09:35:00 +0000 (09:35 +0000)]
[LINUX] CONFIG_XEN_BACKEND=m build properly.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
19 years ago[LINUX] Make CONFIG_XEN_FRAMEBUFFER=m build properly.
kfraser@localhost.localdomain [Fri, 8 Dec 2006 09:34:24 +0000 (09:34 +0000)]
[LINUX] Make CONFIG_XEN_FRAMEBUFFER=m build properly.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoMake XEN_DOMCTL_getdomaininfo return shared_info_frame in gmfn not mfn.
kfraser@localhost.localdomain [Fri, 8 Dec 2006 09:32:42 +0000 (09:32 +0000)]
Make XEN_DOMCTL_getdomaininfo return shared_info_frame in gmfn not mfn.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years ago[LINUX] Make xb_waitq static.
kfraser@localhost.localdomain [Fri, 8 Dec 2006 09:27:39 +0000 (09:27 +0000)]
[LINUX] Make xb_waitq static.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
19 years ago[LINUX] Clean up and clarify find_unbound_irq().
kfraser@localhost.localdomain [Fri, 8 Dec 2006 09:19:37 +0000 (09:19 +0000)]
[LINUX] Clean up and clarify find_unbound_irq().
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[IA64] allow foreign domain page mapping of DOMID_XEN/DOMID_IO
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 22:02:37 +0000 (15:02 -0700)]
[IA64] allow foreign domain page mapping of DOMID_XEN/DOMID_IO

DOMID_XEN is used by xenmon, xentrace.

Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years ago[IA64] xc_ia64_linux_restore fix for new foreign domain page mapping semantics
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 22:00:49 +0000 (15:00 -0700)]
[IA64] xc_ia64_linux_restore fix for new foreign domain page mapping semantics

Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years ago[IA64] Domain save fix for foreign domain page mapping semantics change
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 21:56:01 +0000 (14:56 -0700)]
[IA64] Domain save fix for foreign domain page mapping semantics change

Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years ago[IA64] Map vcpu memory mapped registers into guest psudo physical address space
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 21:53:47 +0000 (14:53 -0700)]
[IA64] Map vcpu memory mapped registers into guest psudo physical address space

Assign vcpu memory mapped registers into guest psudo physical address space
for foreign domain page mapping. This is used by domain save.

Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years ago[IA64] Map shared_info into guest pseudo physical address space
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 21:49:00 +0000 (14:49 -0700)]
[IA64] Map shared_info into guest pseudo physical address space

Allows libxc to map from dom0. This is used by domain builder.

Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
19 years agomerge with xen-unstable.hg
awilliam@xenbuild2.aw [Fri, 8 Dec 2006 16:09:18 +0000 (09:09 -0700)]
merge with xen-unstable.hg

19 years ago[IA64] workaround NaT consumption fault when hiding a PCI function from dom0
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 23:18:55 +0000 (16:18 -0700)]
[IA64] workaround NaT consumption fault when hiding a PCI function from dom0

The free_irq() path seems to have trouble with devices hidden from
dom0 for driver domains.  The result is a NaT consumption fault in
the reboot notifier_call_chain() when dom0 reboots.  This workaround
prevents the problem until it can be further investigated.

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years agoFix PERROR.
Ewan Mellor [Thu, 7 Dec 2006 17:48:35 +0000 (17:48 +0000)]
Fix PERROR.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoDiagnose bad config files.
Ewan Mellor [Thu, 7 Dec 2006 17:47:43 +0000 (17:47 +0000)]
Diagnose bad config files.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoMerge
Alastair Tse [Thu, 7 Dec 2006 16:48:23 +0000 (16:48 +0000)]
Merge

19 years ago[XEND] Only return valid VIF and VBD parameters via Xen API.
Alastair Tse [Thu, 7 Dec 2006 16:43:22 +0000 (16:43 +0000)]
[XEND] Only return valid VIF and VBD parameters via Xen API.

Also fix typo for memory_dynamic_{min,max}.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[LINUX] Fail gracefully if we run out of spare IRQs.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 16:43:08 +0000 (16:43 +0000)]
[LINUX] Fail gracefully if we run out of spare IRQs.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoMerge
Alastair Tse [Thu, 7 Dec 2006 16:09:04 +0000 (16:09 +0000)]
Merge

19 years ago[XEND] Sync Xen API with newer VDB and VIF specification.
Alastair Tse [Thu, 7 Dec 2006 16:08:44 +0000 (16:08 +0000)]
[XEND] Sync Xen API with newer VDB and VIF specification.

Fix breakage with storing VIF and VBD refs, remove extraneous
debugging. Also fixing some type errors in default values for Xen API
if things are missing in XendConfig.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[XEN] Add new safe pte update macros and use as necessary when
kfraser@localhost.localdomain [Thu, 7 Dec 2006 14:57:14 +0000 (14:57 +0000)]
[XEN] Add new safe pte update macros and use as necessary when
updating active pagetables.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Make SXP parsing re-use device_add in XendConfig
Alastair Tse [Thu, 7 Dec 2006 14:40:13 +0000 (14:40 +0000)]
[XEND] Make SXP parsing re-use device_add in XendConfig

This will fix the IP address configuration problem for vifs.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[XEND] Prevent None values from appearing in Device SXP
Alastair Tse [Thu, 7 Dec 2006 14:38:16 +0000 (14:38 +0000)]
[XEND] Prevent None values from appearing in Device SXP

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[XEN] Early page fault handler to deal with spurious page faults.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 14:03:54 +0000 (14:03 +0000)]
[XEN] Early page fault handler to deal with spurious page faults.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Fix resume for managed domains.
Alastair Tse [Thu, 7 Dec 2006 13:47:13 +0000 (13:47 +0000)]
[XEND] Fix resume for managed domains.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years agoMerge
Tim Deegan [Thu, 7 Dec 2006 13:18:08 +0000 (13:18 +0000)]
Merge

19 years ago[XEN] Init pae_l3_cache lock earlier
Tim Deegan [Thu, 7 Dec 2006 13:14:44 +0000 (13:14 +0000)]
[XEN] Init pae_l3_cache lock earlier
In particular, before hvm_vcpu_initialise causes a CR3 update.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[IA64] Send events to VTI domain through level triggered irq.
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 12:43:04 +0000 (05:43 -0700)]
[IA64] Send events to VTI domain through level triggered irq.

Signed-off-by: Anthony Xu <anthony.xu@intel.com>
19 years ago[IA64] fix warnings
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 12:36:30 +0000 (05:36 -0700)]
[IA64] fix warnings

Signed-off-by: Anthony Xu <anthony.xu@intel.com>
19 years ago[IA64] Implement irq redirection of IOSAPIC
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 12:34:07 +0000 (05:34 -0700)]
[IA64] Implement irq redirection of IOSAPIC

Signed-off-by: Anthony Xu <anthony.xu@intel.com>
19 years agoThis patch checks the contents of XENDOMAINS_SAVE before the restore
Ewan Mellor [Thu, 7 Dec 2006 12:23:35 +0000 (12:23 +0000)]
This patch checks the contents of XENDOMAINS_SAVE before the restore
process begins, and prevents xendomains from attempting to start any
domain that appears there, whether the domain started successfully or
not.

Signed off by: Hugh Brock <hbrock@redhat.com>

19 years agoThe new lifecycle patches in xen-unstable added 3 new operations which can
Ewan Mellor [Thu, 7 Dec 2006 12:15:30 +0000 (12:15 +0000)]
The new lifecycle patches in xen-unstable added 3 new operations which can
be performed on domains - new, start and delete.  libvirt has had no-op
stub implementations of these APIs for a while & thus we'd enable their
use for Xen 3.0.4 builds. The new APIs are not currently exposed via the
existing SEXPR protocol, and since the new Xen-API protocol is unsupported
tech-preview for 3.0.4 we don't want to re-write against that just yet.

Thus the attached patch adds the 3 neccessary bindings for the SEXPR protocol
to allow the lifecycle functionality to be invoked. It also allows the 'state'
parameter to be specified when listing domains, so we can explicitly request
inactive domains if we know we're running against a 3.0.4 era XenD

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
19 years agoScrub VNC passwords from the logs.
Ewan Mellor [Thu, 7 Dec 2006 12:14:22 +0000 (12:14 +0000)]
Scrub VNC passwords from the logs.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
19 years agoDo an explicit check for bootloader path existence prior
Ewan Mellor [Thu, 7 Dec 2006 12:11:40 +0000 (12:11 +0000)]
Do an explicit check for bootloader path existence prior
to the one for executable permissions. Also print out the path being
tested.

Signed-off-by: Daniel P, Berrange <berrange@redhat.com>
19 years agoThe HTTP service used for SEXPR based RPC calls currently serializes all
Ewan Mellor [Thu, 7 Dec 2006 12:07:53 +0000 (12:07 +0000)]
The HTTP service used for SEXPR based RPC calls currently serializes all
incoming client requests. Since some requests can take non-trivial time
blocking out other clients in this way has bad consequences. The attached
patch makes XenD spawn a new thread to handle each incoming HTTP request.
This same approach is already taken in the XML-RPC service in XenD so I
don't anticipate any multi-thread issues (at least none which don't already
exist).

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
19 years ago[HVM] qemu mouse: Adds support for USB mouse/tablet status check and
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:52:26 +0000 (11:52 +0000)]
[HVM] qemu mouse: Adds support for USB mouse/tablet status check and
restricts Universal Host Controller interrupt generating when received
NAK in interrupt transfer.

According to usb spec, USB mouse/tablet device returns NAK to host
controller if its status does not alter in interrupt transfer.
And UHC should leave a TD active when receiving NAK and execute this
incompleted TD in a subseqent frame. UHC only generates an interrupt
on complete after the TD with ICO bit is completed.

This patch make UHC & USB mouse/tablet behave consistently with spec.

Signed-off-by: Xinmei Huang <xinmei.huang@intel.com>
19 years agoMerge
Alastair Tse [Thu, 7 Dec 2006 11:51:22 +0000 (11:51 +0000)]
Merge

19 years ago[XM] Fix xm uptime TypeError exception.
Alastair Tse [Thu, 7 Dec 2006 11:50:11 +0000 (11:50 +0000)]
[XM] Fix xm uptime TypeError exception.

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
19 years agomerge
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:47:10 +0000 (11:47 +0000)]
merge

19 years ago[LIBXC] Python error extraction tweak.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:46:07 +0000 (11:46 +0000)]
[LIBXC] Python error extraction tweak.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoRemove broken call to XendDomainInfo.getSsidref(), fixing the obsolete HTTP
Ewan Mellor [Thu, 7 Dec 2006 11:45:04 +0000 (11:45 +0000)]
Remove broken call to XendDomainInfo.getSsidref(), fixing the obsolete HTTP
server.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoMerge
Alastair Tse [Thu, 7 Dec 2006 11:45:00 +0000 (11:45 +0000)]
Merge

19 years agoMerge
Alastair Tse [Thu, 7 Dec 2006 11:44:05 +0000 (11:44 +0000)]
Merge

19 years agoRemove unused KEXEC_RANGE_VA_XEN.
Ian Campbell [Thu, 7 Dec 2006 11:43:19 +0000 (11:43 +0000)]
Remove unused KEXEC_RANGE_VA_XEN.

KEXEC_RANGE_VA_XEN is currently unused and there is no point in exporting it.

Signed-Off-By: Magnus Damm <magnus@valinux.co.jp>
---

 Applies to xen-unstable-12756

 xen/common/kexec.c         |   13 +++----------
 xen/include/public/kexec.h |    3 +--
 2 files changed, 4 insertions(+), 12 deletions(-)

19 years ago[XEND] Fix configureBootloader to work with HTTP/SEXP.
Alastair Tse [Thu, 7 Dec 2006 11:41:38 +0000 (11:41 +0000)]
[XEND] Fix configureBootloader to work with HTTP/SEXP.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
19 years ago[LIBXC] Add an error reporting API to the libxc library.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:36:26 +0000 (11:36 +0000)]
[LIBXC] Add an error reporting API to the libxc library.

 - An 'xc_error' struct is used to pass around error
   details. Currently contains two members 'code' an enumeration of
   error types, and 'message' a free text description of the specific
   problem.

 - The xc_get_last_error() method returns a const pointer to the
   internal instance of this struct manged by libxc. By returning a
   const pointer we can add extra members to the end of the struct at
   any time without worrying about ABI of callers. This will let us
   provide more fine-grained info if needed in the future.

 - The xc_error instance is statically defined inside libxc and marked
   __thread. This ensures that errors are recorded per-thread, and
   that when dealing with errors we never need to call malloc - all
   storage needed is statically allocated.

 - The xc_clear_last_error() method resets any currently recorded
   error details

 - The xc_error_code_to_desc() method converts the integer error code
   into a generic user facing messsage. eg "Invalid kernel". Together
   with the 'message' field from xc_error, this provides the user
   visible feedback. eg "Invalid kernel: Non PAE-kernel on PAE host."

 - A callback can be registered with xc_set_error_handler to receive
   notification whenever an error is recorded, rather than querying
   for error details after the fact with xc_get_last_error

 - If built with -DDEBUG set, a default error handler will be
   registered which calls fprintf(stderr), thus maintaining current
   behaviour of logging errors to stderr during developer builds.

 - The python binding for libxc is updated to use xc_get_last_error
   to pull out error details whenever appropriate, instead of
   returning info based on 'errno'

 - The xc_set_error method is private to libxc internals, and is used
   for setting error details

 - The ERROR and PERROR macros have been updated to call xc_set_error
   automatically specifying XC_INTERNAL_ERROR as the error code. This
   gives a generic error report for all current failure points

 - Some uses of the ERROR macro have been replaced with explicit
   calls to xc_set_error to enable finer grained error reporting. In
   particular the code dealing with invalid kernel types uses this
   to report about PAE/architecture/wordsize mismatches

The patch has been tested by calling xm create against a varietry of
config files defining invalid kernels of various kinds. It has also
been tested with libvirt talking to xend. In both cases the error
messages were propagated all the way back up the stack.

There is only one place where I need to do further work. The suspend
& restore APIs in Xend invoke external helper programs rather than
calling libxc directly. This means that error details are essentially
lost. Since there is already code in XenD which scans STDERR from
these programs I will investigate adapting this to extract actual
error messages from these helpers.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
19 years ago[XEN] Make multiboot-related code more readable.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:22:26 +0000 (11:22 +0000)]
[XEN] Make multiboot-related code more readable.
Tested on i386 and x86_64 with GRUB and PXELINUX bootmanagers.

From: Christoph Egger <Christoph.Egger@amd.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoVMXAssist cleanups
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:18:38 +0000 (11:18 +0000)]
VMXAssist cleanups

Signed-off-by: Xin Li <xin.b.li@intel.com>
19 years ago[VTPM][HVM] Transmit the locality in which the command was sent in the
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:17:28 +0000 (11:17 +0000)]
[VTPM][HVM] Transmit the locality in which the command was sent in the
highest 3 bits of the 4 bytes prepended to each packet.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
19 years ago[HVM] Fix reboot failure when configured 'cdrom', i.e. bug 832.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:16:17 +0000 (11:16 +0000)]
[HVM] Fix reboot failure when configured 'cdrom', i.e. bug 832.
Signed-off-by : Xinmei Huang <xinmei.huang@intel.com>

19 years ago[IA64] implement free_irq_vector() paravirtualization
awilliam@xenbuild2.aw [Thu, 7 Dec 2006 11:15:54 +0000 (04:15 -0700)]
[IA64] implement free_irq_vector() paravirtualization

This avoids a double free error message since dom0 isn't
actually using the ia64_vector_mask.

Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
19 years agomerge
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:14:21 +0000 (11:14 +0000)]
merge

19 years ago[HVM] qemu: Add guest address-space mapping cache.
kfraser@localhost.localdomain [Thu, 7 Dec 2006 11:12:52 +0000 (11:12 +0000)]
[HVM] qemu: Add guest address-space mapping cache.

On IA32 host or IA32 PAE host, at present, generally, we can't create
an HVM guest with more than 2G memory, because generally it's almost
impossible for Qemu to find a large enough and consecutive virtual
address space to map an HVM guest's whole physical address space.
The attached patch fixes this issue using dynamic mapping based on
little blocks of memory.

Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>