xen.git
19 years ago[POWERPC][XEN] Better explane the larx/stcx issue with interrupts
Jimi Xenidis [Thu, 14 Sep 2006 14:17:23 +0000 (10:17 -0400)]
[POWERPC][XEN] Better explane the larx/stcx issue with interrupts

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Safety with foreign get_page() calls and RMA
Jimi Xenidis [Thu, 14 Sep 2006 05:41:13 +0000 (01:41 -0400)]
[POWERPC][XEN] Safety with foreign get_page() calls and RMA

The following patch deals with get_page() possibly failing for H_ENTER
on a foreign page and returning the correct error.  We also tag and
checke that a put_page() for RMA pages will panic (for now) if the
domain is _not_ dying.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Inline clear_page() and use DCBZ to optimize
Jimi Xenidis [Thu, 14 Sep 2006 05:36:39 +0000 (01:36 -0400)]
[POWERPC][XEN] Inline clear_page() and use DCBZ to optimize

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Define all cache information globaly
Jimi Xenidis [Thu, 14 Sep 2006 05:12:26 +0000 (01:12 -0400)]
[POWERPC][XEN] Define all cache information globaly

Rather than use the config.h macro

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Track pages correctly
Jimi Xenidis [Wed, 13 Sep 2006 22:41:11 +0000 (18:41 -0400)]
[POWERPC][XEN] Track pages correctly

The following patch tracks and frees pages correctly.  It solves the problem where a foreign mapping would zombie a domain because the refcnts remained.
This involved:
  - implement relinquish_memory() for PowerPC
  - remove free_rma() since all pages get "relinquished" now.
  - foreign pages are get and put correctly

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEN][POWERPC] Split dom0_ops.c into domctl.c and sysctl.c.
Hollis Blanchard [Tue, 12 Sep 2006 19:28:16 +0000 (14:28 -0500)]
[XEN][POWERPC] Split dom0_ops.c into domctl.c and sysctl.c.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Handle foreign page mappings correctly
Jimi Xenidis [Tue, 12 Sep 2006 15:01:04 +0000 (11:01 -0400)]
[POWERPC][XEN] Handle foreign page mappings correctly

The following patch performs the following:
  - add free_rma() to go with allocate_rma()
  - allocate_rma() returns -EINVAL if an RMA is already allocated
  - PFN_TYPE_REMOTE is now PFN_TYPE_FOREIGN
  - better checking of page_info in pfn2mfn()
  - Stop using RPN and LPN and use PFN and MFN respectively
  - Increase/decrease foreign domain refcount when another domain
    maps/unmaps it

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] make sure put_domain() is called in case of allocate_rma() failuer
Jimi Xenidis [Tue, 12 Sep 2006 14:53:46 +0000 (10:53 -0400)]
[POWERPC][XEN] make sure put_domain() is called in case of allocate_rma() failuer

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] if crash_debug=n then at least dump a backtrace.
Jimi Xenidis [Tue, 12 Sep 2006 10:48:32 +0000 (06:48 -0400)]
[POWERPC][XEN] if crash_debug=n then at least dump a backtrace.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Detect bad spurious interrupt condition and panic instead of hang
Jimi Xenidis [Tue, 12 Sep 2006 10:47:22 +0000 (06:47 -0400)]
[POWERPC][XEN] Detect bad spurious interrupt condition and panic instead of hang

When handing off the MPIC from Xen to Dom0, which is the current yet
not permamnet design, the MPIC can cause the processor to assert an
external interrupt when none is available.  Rather then simply hang in
this condition we now panic so the user can see that there is indeed a
problem and identify it as this one.

This condition seems to be related to temperature and the probablity
of it occuring decreases if the machine is allowed to stay idle (not
in the Xen panic loop) for a minute or two.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEN][POWERPC] fix timebase race in SMP init
Hollis Blanchard [Fri, 8 Sep 2006 17:28:49 +0000 (12:28 -0500)]
[XEN][POWERPC] fix timebase race in SMP init

cpuid was being set twice in cpu_online_map, and the first one was causing a
race in the timebase synchronization code.

Signed-off-by: Amos Waterland <apw@us.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEN][POWERPC] Fix allocation error for xencomm "mini" descriptors.
Hollis Blanchard [Fri, 8 Sep 2006 15:23:56 +0000 (10:23 -0500)]
[XEN][POWERPC] Fix allocation error for xencomm "mini" descriptors.

Previous code worked correctly only if the allocated structure fit in the same
page as the base address. If the structure did not fit, the computation for the
allocation was incorrect. Noticed by Kiyokuni KAWACHIYA <KAWATIYA@jp.ibm.com>.

Signed-off-by: Maria Butrico <butrico@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEN][POWERPC] Fix machine check build break.
Hollis Blanchard [Fri, 8 Sep 2006 15:22:45 +0000 (10:22 -0500)]
[XEN][POWERPC] Fix machine check build break.
Building exceptions.c with CRASH_DEBUG=y produces a build break, because
gcc complains about the unused variable.

Signed-off-by: Amos Waterland <apw@us.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[LIBXC][POWERPC] Add missing xc_memory.c
hollisb@localhost [Mon, 18 Sep 2006 18:08:58 +0000 (13:08 -0500)]
[LIBXC][POWERPC] Add missing xc_memory.c
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[LIBXC][POWERPC] Update to latest flatdevtree snapshot.
Hollis Blanchard [Thu, 7 Sep 2006 22:25:52 +0000 (17:25 -0500)]
[LIBXC][POWERPC] Update to latest flatdevtree snapshot.

Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Track HID4[RM_CI] state for machine check
Jimi Xenidis [Fri, 8 Sep 2006 01:46:33 +0000 (21:46 -0400)]
[POWERPC][XEN] Track HID4[RM_CI] state for machine check

This patch creates a PIR indexed character array so we can note when a
processor is in RM_CI mode.  Machine checks can happen here and we
need to at least report it (and maybe one day) recover from it. Also
some fun with the SCOM.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Include io.h so we can match the globals defined here.
Jimi Xenidis [Thu, 7 Sep 2006 16:09:18 +0000 (12:09 -0400)]
[POWERPC][XEN] Include io.h so we can match the globals defined here.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Synchronize Timebase on all CPUs
Jimi Xenidis [Thu, 7 Sep 2006 06:50:06 +0000 (02:50 -0400)]
[POWERPC][XEN] Synchronize Timebase on all CPUs

This patch add the necessary support to use
  arch/powerpc/kernel/smp-tbsync.c
from Linux.
Original copyright and author are:
  Copyright (C) 2003 Samuel Rydh (samuel@ibrium.se)

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Clear SLB entries on boot and other cleanups
Jimi Xenidis [Thu, 7 Sep 2006 06:21:17 +0000 (02:21 -0400)]
[POWERPC][XEN] Clear SLB entries on boot and other cleanups

This patch clears and SLB entries that might have been left behind by
Firmware and also cleans up the Save and Restore of the segments.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Machine Check Processing
Jimi Xenidis [Thu, 7 Sep 2006 05:48:42 +0000 (01:48 -0400)]
[POWERPC][XEN] Machine Check Processing

This patch allows allows for a processor to report on the cause of the
machine check and possible even recover from it.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] More Robust Memory Checking
Jimi Xenidis [Thu, 7 Sep 2006 05:30:12 +0000 (01:30 -0400)]
[POWERPC][XEN] More Robust Memory Checking

This patch allows the platform to define the "IO Hole" of
addressibility and checks that even Dom0 does not try to Map "Remote"
memory that is not there.  Also replaces some panic calls with WARN();
and returns failure.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] move setting of MSR[RI] till after SRR0/1
Jimi Xenidis [Tue, 5 Sep 2006 19:25:06 +0000 (15:25 -0400)]
[POWERPC][XEN] move setting of MSR[RI] till after SRR0/1

This also frees up space so we can properly/safely blow away larx/stcx
reservations.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Make many of the timbase routine static inline
Jimi Xenidis [Fri, 1 Sep 2006 17:14:53 +0000 (13:14 -0400)]
[POWERPC][XEN] Make many of the timbase routine static inline

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] make sure Dom0.VCPU0 is pinned to the booting CPU
Jimi Xenidis [Fri, 1 Sep 2006 16:37:29 +0000 (12:37 -0400)]
[POWERPC][XEN] make sure Dom0.VCPU0 is pinned to the booting CPU

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] move idle_loop and add a "power save" hook
Jimi Xenidis [Fri, 1 Sep 2006 16:31:56 +0000 (12:31 -0400)]
[POWERPC][XEN] move idle_loop and add a "power save" hook

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] remove dead preemption code
Jimi Xenidis [Fri, 1 Sep 2006 16:19:41 +0000 (12:19 -0400)]
[POWERPC][XEN] remove dead preemption code

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Adjust DEC correctly
Jimi Xenidis [Fri, 1 Sep 2006 16:12:43 +0000 (12:12 -0400)]
[POWERPC][XEN] Adjust DEC correctly

If DEC would have expired we now restore DEC to 0 so it will fire.
The Domain loses info about how late DEC was from looking at DEC but I
don't think anyone tracks that.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Erratum: Must clear larx/stcx reservation on exception
Jimi Xenidis [Fri, 1 Sep 2006 15:42:51 +0000 (11:42 -0400)]
[POWERPC][XEN] Erratum: Must clear larx/stcx reservation on exception

PowerPC 970 Erratum that an "OS should execute a stcx in
the interrupt handler to clear the reservation."

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC][XEN] Make MPIC sharing between Xen and Dom0 optional
Jimi Xenidis [Thu, 31 Aug 2006 21:47:11 +0000 (17:47 -0400)]
[POWERPC][XEN] Make MPIC sharing between Xen and Dom0 optional

Tho right now it is the only option.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[POWERPC] merge with xen-unstable.hg
hollisb@localhost [Mon, 18 Sep 2006 17:48:56 +0000 (12:48 -0500)]
[POWERPC] merge with xen-unstable.hg
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years agoMerge with xenppc-unstable-merge.hg.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 21:36:18 +0000 (22:36 +0100)]
Merge with xenppc-unstable-merge.hg.

19 years agoMerge with xen-ia64-unstable.hg
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 20:38:34 +0000 (21:38 +0100)]
Merge with xen-ia64-unstable.hg

19 years ago[POWERPC] merge with xen-unstable.hg
Hollis Blanchard [Wed, 30 Aug 2006 19:09:31 +0000 (14:09 -0500)]
[POWERPC] merge with xen-unstable.hg

19 years ago[NET] front: move register_netdev() + sysfs attribute registering from create_netdev...
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 17:40:40 +0000 (18:40 +0100)]
[NET] front: move register_netdev() + sysfs attribute registering from create_netdev() into the new open_netdev() function.
open_netdev is called as very last thing in netdev_probe.

This closes at least one initialization race: uninitialized
mac address is visible for a short time in
/sys/class/net/${if}/address

Signed-off-by: Gerd Hoffmann <kraxel@suse.de>
19 years ago[XEN] Add a start_pfn field in xen_domctl_getmemlist.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 17:19:04 +0000 (18:19 +0100)]
[XEN] Add a start_pfn field in xen_domctl_getmemlist.
This is used by ia64 because it directly reads the P2M table.
Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
19 years agoRevert powerpc build break after dom0_ops split.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 17:08:19 +0000 (18:08 +0100)]
Revert powerpc build break after dom0_ops split.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEN] Reserve hypercall number for kexec operations.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 17:03:55 +0000 (18:03 +0100)]
[XEN] Reserve hypercall number for kexec operations.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] physdev: Check for negative irqs in alloc_irq_vector.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 17:02:32 +0000 (18:02 +0100)]
[XEN] physdev: Check for negative irqs in alloc_irq_vector.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
19 years ago[LINUX][X86_64] P2M map must be allocated on page aligned boundary.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 16:56:49 +0000 (17:56 +0100)]
[LINUX][X86_64] P2M map must be allocated on page aligned boundary.
Otherwise save/restore can break.
Signed-off-by: Kazuo Moriwaka <moriwaka@valinux.co.jp>
19 years ago[TOOLS] Fix gdbserver build.
kaf24@firebug.cl.cam.ac.uk [Wed, 30 Aug 2006 16:52:35 +0000 (17:52 +0100)]
[TOOLS] Fix gdbserver build.
Signed-off-by: Travis Betak <travis.betak@amd.com>
19 years agoFix fix :-)
Steven Hand [Wed, 30 Aug 2006 12:51:12 +0000 (13:51 +0100)]
Fix fix :-)

Signed-off-by: Steven Hand <steven@xensource.com>
19 years agoFix hard tabs.
Steven Hand [Wed, 30 Aug 2006 12:50:08 +0000 (13:50 +0100)]
Fix hard tabs.

Signed-off-by: Steven Hand <steven@xensource.com>
19 years agoBlock IO safety patch for live migration. Xend checks that VBDs are destroyed before...
jchesterfield@dhcp92.uk.xensource.com [Wed, 30 Aug 2006 10:39:27 +0000 (11:39 +0100)]
Block IO safety patch for live migration. Xend checks that VBDs are destroyed before unpausing the domain.

19 years agoFurther shadow_mem_control fixes -- we need to balloon the amount of memory
Ewan Mellor [Wed, 30 Aug 2006 10:05:00 +0000 (11:05 +0100)]
Further shadow_mem_control fixes -- we need to balloon the amount of memory
that we're going to request i.e. the rounded-up value.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoShadow memory should be in MiB, not KiB. It also needs rounding up to be safe.
Ewan Mellor [Wed, 30 Aug 2006 08:47:24 +0000 (09:47 +0100)]
Shadow memory should be in MiB, not KiB.  It also needs rounding up to be safe.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoFix reboot and reconfigure of HVM guests when they are configured with an
Ewan Mellor [Wed, 30 Aug 2006 01:53:48 +0000 (02:53 +0100)]
Fix reboot and reconfigure of HVM guests when they are configured with an
empty CD-ROM drive.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoAdded an entry to arch._types for x86_64.
Ewan Mellor [Wed, 30 Aug 2006 00:20:01 +0000 (01:20 +0100)]
Added an entry to arch._types for x86_64.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoFix typo.
Andrew Warfield [Tue, 29 Aug 2006 23:31:57 +0000 (16:31 -0700)]
Fix typo.

Signed-off-by: <andy@xensource.com>
19 years agomerge
Andrew Warfield [Tue, 29 Aug 2006 22:58:57 +0000 (15:58 -0700)]
merge

19 years agoSmall blktap cleanups.
akw@devbox.site [Tue, 29 Aug 2006 22:55:24 +0000 (15:55 -0700)]
Small blktap cleanups.

Signed-off-by: Andrew Warfield <andy@xensource.com>
19 years ago[XEN][POWERPC] catch up with xen-unstable changes
Hollis Blanchard [Tue, 29 Aug 2006 22:53:57 +0000 (17:53 -0500)]
[XEN][POWERPC] catch up with xen-unstable changes
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEND] abstract architecture-specific bits in image.py
Ewan Mellor [Tue, 29 Aug 2006 22:20:22 +0000 (23:20 +0100)]
[XEND] abstract architecture-specific bits in image.py
- create arch.type (which evaluates to "x86", "ia64" or "powerpc")
- create subclasses for x86 and ia64 HVM loaders
- rework findImageHandlerClass()
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years agoThis patch fixed 'xm reboot' to work as you expected with the wait
Ewan Mellor [Tue, 29 Aug 2006 21:51:06 +0000 (22:51 +0100)]
This patch fixed 'xm reboot' to work as you expected with the wait
option. This patch adds a restart sequence counter to xenstore.
The completion of a VM reboot is detected by observing that
its restart sequence counter changes.

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
19 years agoImprove xm migrate error message by printing usage instead of "Invalid
Ewan Mellor [Tue, 29 Aug 2006 21:51:05 +0000 (22:51 +0100)]
Improve xm migrate error message by printing usage instead of "Invalid
arguments".

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
19 years agoReduce AIO context allocations in blktap.
akw@devbox.site [Tue, 29 Aug 2006 21:34:57 +0000 (14:34 -0700)]
Reduce AIO context allocations in blktap.

Signed-off-by: Andrew Warfield <andy@xensource.com>
19 years ago[IA64] revert xen-ia64-unstable.hg cset 11301:a19dbbe4cff5
awilliam@xenbuild.aw [Tue, 29 Aug 2006 15:08:29 +0000 (09:08 -0600)]
[IA64] revert xen-ia64-unstable.hg cset 11301:a19dbbe4cff5

revert arch_domain_destory() for old PV-on-HVM on IPF

Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Tomonari Horikoshi <t.horikoshi@jp.fujitsu.com>
19 years ago[IA64] revert xen-ia64-unstable.hg 11271:c232365128cf
awilliam@xenbuild.aw [Tue, 29 Aug 2006 15:05:46 +0000 (09:05 -0600)]
[IA64] revert xen-ia64-unstable.hg 11271:c232365128cf

Delete functions introduced for old PV-on-HVM on IPF.

Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Tomonari Horikoshi <t.horikoshi@jp.fujitsu.com>
19 years agomerge with xen-unstable.hg
awilliam@xenbuild.aw [Tue, 29 Aug 2006 14:28:38 +0000 (08:28 -0600)]
merge with xen-unstable.hg

19 years ago[IA64] init the hlt timer on a valid cpu and migrate
awilliam@xenbuild.aw [Tue, 29 Aug 2006 14:09:28 +0000 (08:09 -0600)]
[IA64] init the hlt timer on a valid cpu and migrate

Based on a patch from Anthony Xu, v->processor is not initialized
in alloc_vcpu_struct() when we initialize the hlt_timer.  Init the
timer on the first cpu and migrate it in schedule_tail().

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years agoFix definition of LINUX_VER so that doesn't pickup LINUX_VER3
Ian Campbell [Tue, 29 Aug 2006 11:23:11 +0000 (12:23 +0100)]
Fix definition of LINUX_VER so that doesn't pickup LINUX_VER3
from mk.linux-2.6-xen. This makes "make mkpatches" work again.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[XEN][POWERPC] cleanup for (;;);
Jimi Xenidis [Tue, 29 Aug 2006 10:53:58 +0000 (06:53 -0400)]
[XEN][POWERPC] cleanup for (;;);

When the port first started, we didn't even have panic() working
properly so we used "for(;;);" loops to stop execution.  Now that output
is buffered these look like hangs, since you usually do not see the
printk() before.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years agoMerge
tdeegan@york.uk.xensource.com [Tue, 29 Aug 2006 08:42:35 +0000 (09:42 +0100)]
Merge

19 years ago[XEN] Off-by one error in range checks translating gfns to mfns
tdeegan@york.uk.xensource.com [Tue, 29 Aug 2006 08:37:30 +0000 (09:37 +0100)]
[XEN] Off-by one error in range checks translating gfns to mfns
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years agomerge with xen-unstable.hg
awilliam@xenbuild.aw [Mon, 28 Aug 2006 22:26:37 +0000 (16:26 -0600)]
merge with xen-unstable.hg

19 years ago[IA64] Modify sparse headers for PV-on-HVM/IPF
awilliam@xenbuild.aw [Mon, 28 Aug 2006 22:16:07 +0000 (16:16 -0600)]
[IA64] Modify sparse headers for PV-on-HVM/IPF

Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Tomonari Horikoshi <t.horikoshi@jp.fujitsu.com>
19 years ago[IA64] separate hypercalls, acpi and EFI tables
awilliam@xenbuild.aw [Mon, 28 Aug 2006 22:09:28 +0000 (16:09 -0600)]
[IA64] separate hypercalls, acpi and EFI tables

Put hypercalls, acpi and EFI tables on three separate 4KB pages.
This was a stopper to use 4KB pages.
Comments added and simplifications.

Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
19 years ago[qemu patches] Update patches for changeset 11273:a1cff03ac7d7.
Christian Limpach [Mon, 28 Aug 2006 21:44:31 +0000 (22:44 +0100)]
[qemu patches] Update patches for changeset 11273:a1cff03ac7d7.

Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
19 years ago[qemu] Fix PCI config header space writes and set Subsystem-Vendor ID fields.
Christian Limpach [Mon, 28 Aug 2006 21:43:09 +0000 (22:43 +0100)]
[qemu] Fix PCI config header space writes and set Subsystem-Vendor ID fields.
Fix failure in Windows HCT's PCI-Compliance-Test by invalidating the
writing to read-only/reserved fields in PCI configuration space header
and giving Subsystem-Vendor-ID field a valid value via copying Vendor-ID
field into it.

Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
19 years ago[XEN][POWERPC] do not set d->tot_pages, but check it.
Jimi Xenidis [Mon, 28 Aug 2006 21:28:55 +0000 (17:28 -0400)]
[XEN][POWERPC] do not set d->tot_pages, but check it.
Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years agoFix hypervisor build with performance counters after -unstable
Steven Hand [Mon, 28 Aug 2006 20:57:34 +0000 (21:57 +0100)]
Fix hypervisor build with performance counters after -unstable
cset 11257:86d26e6ec89b

Signed-off-by: Steve Dobbelstein <steved@us.ibm.com>
19 years agoFix shadow mode 32/PAE linux writable heuristic and add 64-bit version.
Steven Hand [Mon, 28 Aug 2006 20:40:42 +0000 (21:40 +0100)]
Fix shadow mode 32/PAE linux writable heuristic and add 64-bit version.

Signed-off-by: Steven Hand <steven@xensource.com>
19 years agoFix PAE-mode HVM guests with the new shadow code.
Steven Hand [Mon, 28 Aug 2006 20:25:13 +0000 (21:25 +0100)]
Fix PAE-mode HVM guests with the new shadow code.

Signed-off-by: Steven Hand <steven@xensource.com>
19 years ago[IA64] fix noreboot option
awilliam@xenbuild.aw [Mon, 28 Aug 2006 20:09:31 +0000 (14:09 -0600)]
[IA64] fix noreboot option

fix machine_hlt to support noreboot option.

Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
[modified to use existing cpu_halt() function in linux-xen files]
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years ago[IA64] print command line of xen
awilliam@xenbuild.aw [Mon, 28 Aug 2006 19:31:35 +0000 (13:31 -0600)]
[IA64] print command line of xen

Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
19 years ago[IA64] fix warnings
awilliam@xenbuild.aw [Mon, 28 Aug 2006 19:30:04 +0000 (13:30 -0600)]
[IA64] fix warnings

Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
19 years ago[qemu patches] Update patches for changeset 11268:350a73222a51.
Christian Limpach [Mon, 28 Aug 2006 19:22:56 +0000 (20:22 +0100)]
[qemu patches] Update patches for changeset 11268:350a73222a51.

Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
19 years ago[qemu] Add target-i386-dm and hw *.[ch] files to TAGS.
Christian Limpach [Mon, 28 Aug 2006 19:21:38 +0000 (20:21 +0100)]
[qemu] Add target-i386-dm and hw *.[ch] files to TAGS.

Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>
19 years ago[IA64] fix VTi hangs
awilliam@xenbuild.aw [Mon, 28 Aug 2006 19:18:36 +0000 (13:18 -0600)]
[IA64] fix VTi hangs

In some scenarios, xen_timer_interrupt will not set machine itm,
this may make this LP not receive timer interrupt any more.

Signed-off-by: Anthony Xu <anthony.xu@intel.com>
19 years ago[IA64] Remove warning message in fw_emul.c
awilliam@xenbuild.aw [Mon, 28 Aug 2006 19:14:52 +0000 (13:14 -0600)]
[IA64] Remove warning message in fw_emul.c

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
19 years ago[IA64] Fix comment of vmx_hypercall.c
awilliam@xenbuild.aw [Mon, 28 Aug 2006 19:12:42 +0000 (13:12 -0600)]
[IA64] Fix comment of vmx_hypercall.c

Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Tomonari Horikoshi <t.horikoshi@jp.fujitsu.com>
19 years ago[IA64] Modify destroy code for PV-on-HVM on IPF
awilliam@xenbuild.aw [Mon, 28 Aug 2006 19:11:17 +0000 (13:11 -0600)]
[IA64] Modify destroy code for PV-on-HVM on IPF

Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Tomonari Horikoshi <t.horikoshi@jp.fujitsu.com>
19 years ago[HVM] Don't assume that VMENTER causes a TLB flush, when validating
tdeegan@york.uk.xensource.com [Mon, 28 Aug 2006 16:50:21 +0000 (17:50 +0100)]
[HVM] Don't assume that VMENTER causes a TLB flush, when validating
shadow pagetable updates.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[XEN][POWERPC] correct HTAB allocation typo
Jimi Xenidis [Mon, 28 Aug 2006 15:52:17 +0000 (11:52 -0400)]
[XEN][POWERPC] correct HTAB allocation typo
Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years agoFix Intel copyright notice in a single source file.
kaf24@localhost.localdomain [Mon, 28 Aug 2006 12:08:41 +0000 (13:08 +0100)]
Fix Intel copyright notice in a single source file.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Move all of the various log files created by xen to
kaf24@localhost.localdomain [Mon, 28 Aug 2006 12:05:00 +0000 (13:05 +0100)]
[XEND] Move all of the various log files created by xen to
be under /var/log/xen instead of under /var/log directly.  This has
the advantage of cleaning things up a little and also can make it
easier to restrict the permissions needed by xend.

Signed-off-by: Jeremy Katz <katzj@redhat.com>
19 years ago[HVM] [SMBIOS] Duplicate UUID into serial number in SMBIOS tables.
kaf24@localhost.localdomain [Mon, 28 Aug 2006 11:59:07 +0000 (12:59 +0100)]
[HVM] [SMBIOS] Duplicate UUID into serial number in SMBIOS tables.

Windows zeroes out the UUID in the SMBIOS tables, but the identifier
is often needed by systems management code.  Duplicating the UUID
as the virtual system's serial number should help.

Signed-off-by: Andrew D. Ball <aball@us.ibm.com>
19 years agoThe following patch fixes a bug where xenconsoled will can SEGV
kaf24@localhost.localdomain [Mon, 28 Aug 2006 11:50:55 +0000 (12:50 +0100)]
The following patch fixes a bug where xenconsoled will can SEGV
because it uses FD_ISSET(-1,xxx).  Since the code is written that any
ring/tty handler can set d->tty_fd to -1 it has to be checked
_every_time_.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
19 years ago[TOOLS] Remove Gentoo-specific workaround in networking
kaf24@localhost.localdomain [Mon, 28 Aug 2006 11:49:14 +0000 (12:49 +0100)]
[TOOLS] Remove Gentoo-specific workaround in networking
scripts -- apparently it is not needed (reported by
Tuan Van <lang.thang@gmail.com>). See bugzilla issue #741
for more details.

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[TOOLS] Clean up routed vifs quietly on domain destruction.
kaf24@localhost.localdomain [Mon, 28 Aug 2006 11:41:15 +0000 (12:41 +0100)]
[TOOLS] Clean up routed vifs quietly on domain destruction.

I have experienced that the vif-route script does not work as expected
when deleting a virtual interface (Redhat FC5). Both of the commands
"ifdown" and "ip route" cause an early vif-route script exit and
therefore will cause skipping of the later script commands (such as
cleaning up iptables entries for the default 'antispoof'). The
vif-route
script creates the following syslog error entry:
"/etc/xen/scripts/vif-route failed; error detected."

It appears that both of the problematic commands are actually
redundant
when destroying domains:
1. the interface is already gone (I assume because the domain frontend
is gone already)  --> ifdown does not do anything
2. the route is gone as well since the interface has disappeared -->
ip
route del does not do anything

Executing those redundant commands with "do_without_error" ensures
that
the script completes and cleans up iptables rules. The attached
RFC-patch only masks those commands when bringing down an interface,
so
that domain creation continues to fail in case of vif setup problems
(intended behavior).

Having the iptables cleanup called correctly by vif-route is important
to keep the iptables rule-set clean, otherwise antispoof rules
accumulate with every vif creation.

Signed-off by: Reiner Sailer <sailer@us.ibm.com>

19 years ago[HVM][SVM] Allow SVM to take advantage of flag_dr_dirty.
kaf24@localhost.localdomain [Mon, 28 Aug 2006 11:35:43 +0000 (12:35 +0100)]
[HVM][SVM] Allow SVM to take advantage of flag_dr_dirty.

Signed-off-by: Travis Betak <travis.betak@amd.com>
19 years ago[XEN] Rename shadow2 to shadow and move the various source
kaf24@localhost.localdomain [Mon, 28 Aug 2006 11:09:36 +0000 (12:09 +0100)]
[XEN] Rename shadow2 to shadow and move the various source
files into a sensible directory hierarchy.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN][POWERPC] allocate HTAB using shadow calls
Jimi Xenidis [Mon, 28 Aug 2006 00:48:06 +0000 (20:48 -0400)]
[XEN][POWERPC] allocate HTAB using shadow calls

This patch will use the shadow_ops call to allocate the domains Hashed
Page-table.  This allows for the management tools to customize its size
for the domain.

Warning! The management tools have yet to be updated to use this
properly and update the devtree accordingly so code will assume 64M
RAM and 1M HTAB while the shadow up continues to receive a 0 request.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[XEN][POWERPC] Allocate more memory than RMA for Dom0
Jimi Xenidis [Sun, 27 Aug 2006 20:12:00 +0000 (16:12 -0400)]
[XEN][POWERPC] Allocate more memory than RMA for Dom0

The following patch will only effect Dom0 at the moment and allow it
to be created with memory larger than RMA.  It works by allocating
extents (AKA chunks) of memory of an order specified by the processor.
The extent size should be a value that is capabable of mapping a
"large page" so we use the 970s large page size.  These extents are
tracked by an extents list that simply tracks these page allocations
and what there mapping is.  Other sub-systems effected by this:
  - Dom0 OFD devtree /memory cleanup
  - log_large_page_sizes is now an array
  - Config for minumum Dom0 size

Caveats:
 - There is a hack around HTAB allocation because we currently
   allocate one at domain creation time, this will be correct by a
   patch to follow.
 - Dom0, or domains capable of mapping mfns byt have their extents be
   PFN=MFN

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
19 years ago[IA64] Fixes for dom0_ops changes
awilliam@xenbuild.aw [Sun, 27 Aug 2006 16:25:39 +0000 (10:25 -0600)]
[IA64] Fixes for dom0_ops changes

Fix typo and update op names

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years agomerge with xen-unstable.hg
awilliam@xenbuild.aw [Sun, 27 Aug 2006 16:24:41 +0000 (10:24 -0600)]
merge with xen-unstable.hg

19 years ago[PATCH] fix null pointer dereference in xen_guest_lookup()
Alex Williamson [Sun, 27 Aug 2006 05:56:01 +0000 (06:56 +0100)]
[PATCH] fix null pointer dereference in xen_guest_lookup()

  The latest ELF changes have a null pointer dereference bug when you
have neither an elf notes section nor a __xen_guest_string.  This patch
checks for the existence of the __xen_guest_string prior to using it.
Please apply.  Thanks,

Alex

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
---

19 years ago[IA64] replace lost EFI_RESERVED_TYPE dom0 passthrough
awilliam@xenbuild.aw [Fri, 25 Aug 2006 22:21:39 +0000 (16:21 -0600)]
[IA64] replace lost EFI_RESERVED_TYPE dom0 passthrough

Some systems (HP Superdome in particular) place ACPI tables in
reserved memory regions, so we have to pass it through to dom0.

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years ago[IA64] dom_fw.c cleaned up.
awilliam@xenbuild.aw [Fri, 25 Aug 2006 21:52:31 +0000 (15:52 -0600)]
[IA64] dom_fw.c cleaned up.

Only use one page for stubs and tables instead of 2MB.
Create struct fw_tables to make code more understandable.
Many comments added.
Correctly compute sal systab checksum.

Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>
19 years ago[XEN] Fix silly transposition typo in previous changeset.
kaf24@firebug.cl.cam.ac.uk [Fri, 25 Aug 2006 21:35:24 +0000 (22:35 +0100)]
[XEN] Fix silly transposition typo in previous changeset.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[IA64] Expand hvm_op hypercall for PV-on-HVM/IPF
awilliam@xenbuild.aw [Fri, 25 Aug 2006 21:06:18 +0000 (15:06 -0600)]
[IA64] Expand hvm_op hypercall for PV-on-HVM/IPF

Introduce HVMOP_setup_shared_info_page
 - A page allocated on HVM-guest OS is swapped original shared_info
   page with this hypercall.
 - In x86 code, original shared_info page is used after pv-on-hvm
   setup with remapping feature in arch depend HYPERVISOR_memory_op.
   But, we can't implement same feature for IPF, thus we select to
   implement with this method.
Introduce HVMOP_setup_gnttab_table
 - Pages allocated on HVM-guest OS is swapped original grant_table
   page frames with this hypercall.
 - Same above.

Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Tomonari Horikoshi <t.horikoshi@jp.fujitsu.com>