kfraser@localhost.localdomain [Mon, 11 Jun 2007 14:44:48 +0000 (15:44 +0100)]
Add suspend/resume to devices owned by Xen.
Signed-off-by: Ke Yu <ke.yu@intel.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 11 Jun 2007 14:10:23 +0000 (15:10 +0100)]
x86: Clean up interface to platform timers to extract more common code.
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Mon, 11 Jun 2007 13:56:50 +0000 (14:56 +0100)]
[QEMU] Fix up ioemu timer save/restore after version changes.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Tim Deegan [Mon, 11 Jun 2007 13:38:46 +0000 (14:38 +0100)]
[XEN] Clean up locking/init code around log-dirty interfaces
to avoid deadlocks and make sure locks/functions are in place for
PV domains to be put in log-dirty mode if they're not already shadowed.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Tim Deegan [Mon, 11 Jun 2007 13:35:52 +0000 (14:35 +0100)]
[XEN] Make common log-dirty paging code and add HAP log-dirty support.
Signed-off-by: Wei Huang <wei.huang2@amd.com>
Tim Deegan [Mon, 11 Jun 2007 10:37:10 +0000 (11:37 +0100)]
[TOOLS] sync qemu and python view of save filenames after 15241:
b0109d3dc3dd
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Mon, 11 Jun 2007 09:22:32 +0000 (10:22 +0100)]
xend: Fix duplicate random MAC generation.
Ensure that HVMImageHandler and NetifController see the same random
MAC address.
Signed-off-by: David Edmondson <dme@sun.com>
kfraser@localhost.localdomain [Mon, 11 Jun 2007 09:21:11 +0000 (10:21 +0100)]
xend: Make device detach wait for detach to complete.
*-detach will wait till device is detached or time'ed out.
XendConfig will be updated, if detached.
Signed-off-by: Max Zhen <max.zhen@sun.com>
kfraser@localhost.localdomain [Mon, 11 Jun 2007 09:16:54 +0000 (10:16 +0100)]
xend: Do not give up auto-ballooning if ballooning is proceeding okay.
Under a loaded system, don't give up ballooning dom0 unless it's not
responding at all.
Signed-off-by: Ryan Scott <ryan.scott@sun.com>
kfraser@localhost.localdomain [Mon, 11 Jun 2007 09:14:54 +0000 (10:14 +0100)]
xend: Fix xm block-detach regression introduced by c/s 15157.
Prior to this c/s, user was able to specify device name or id, e.g.
xm block-detach dom xvdb
xm block-detach dom 15728
Now, 'xm block-detach dom xvdb' silently fails. xend.log does contain
the following
[2007-06-07 11:39:18 xend.XendDomainInfo 3775] DEBUG
(XendDomainInfo:519) devName=vbd/51712
[2007-06-07 11:39:18 xend.XendDomainInfo 3775] DEBUG
(XendDomainInfo:527) Could not find the device xvdb
This patch restores the previous behavior but retains the bugfix
introduced by c/s 15571.
Signed-off-by: Jim Fehlig <jfehlig@novell.com>
Keir Fraser [Sun, 10 Jun 2007 18:58:22 +0000 (19:58 +0100)]
tools: Fix xenstored build error by checking vasprintf() return code.
Signed-off-by: Charles Coffing <ccoffing@novell.com>
Keir Fraser [Sun, 10 Jun 2007 18:49:44 +0000 (19:49 +0100)]
hvm: e820 public header cleanup.
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser [Sun, 10 Jun 2007 18:29:32 +0000 (19:29 +0100)]
docs: Update for new 'vga=' boot option syntax.
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser [Sun, 10 Jun 2007 17:07:58 +0000 (18:07 +0100)]
x86: Remove broken and unnecessary numa code from smpboot.c.
This was trampling on random memory when node==NUMA_NO_NODE.
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser [Sun, 10 Jun 2007 16:00:47 +0000 (17:00 +0100)]
x86: Fix save_all regs macro for new CONFIG_FRAME_POINTER define.
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser [Sun, 10 Jun 2007 15:12:51 +0000 (16:12 +0100)]
x86: Fix log-level in printing e820 memory map.
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser [Sun, 10 Jun 2007 14:58:08 +0000 (15:58 +0100)]
x86: Allow selection of graphical video modes during boot.
The 'vga=' boot option is extended as follows:
----------
'vga=<mode-specifier>[,keep]' where <mode-specifier> is one of:
'vga=ask':
display a vga menu of available modes
'vga=text-80x<rows>':
text mode, where <rows> is one of {25,28,30,34,43,50,60}
'vga=gfx-<width>x<height>x<depth>':
graphics mode, e.g., vga=gfx-1024x768x16
'vga=mode-<mode>:
specifies a mode as specified in 'vga=ask' menu
(NB. menu modes are displayed in hex, so mode numbers here must
be prefixed with '0x' (e.g., 'vga=mode-0x0318'))
The option 'keep' causes Xen to continue to print to the VGA console even
after domain 0 starts to boot. The default behaviour is to relinquish
control of the console to domain 0.
----------
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 8 Jun 2007 18:22:08 +0000 (19:22 +0100)]
ioemu: save file name is snprintf()ed into a plenty big enough buffer.
From: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
Steven Hand [Fri, 8 Jun 2007 16:37:58 +0000 (17:37 +0100)]
Avoid using /tmp for qemu state files since an unprivileged user
in dom0 could potentially exploit the fact.
Thanks to Dan Berrange <berrange@redhat.com>
Signed-off-by: Steven Hand
kfraser@localhost.localdomain [Fri, 8 Jun 2007 10:21:21 +0000 (11:21 +0100)]
tools: Fix xm mem-max command.
When I tested xm mem-max and xm list, I saw the following error
message by xm list. Because xm mem-max always changes
"memory_static_max", I think that the cause of the error message
is it. This patch fixes it.
# xm list --long PVdomain.1 | grep mem
(maxmem 1024)
(memory 1024)
(shadow_memory 0)
# xm mem-max PVdomain.1 512
Error: (22, 'Invalid argument')
Usage: xm mem-max <Domain> <Mem>
Set the maximum amount reservation for a domain.
# xm list --long PVdomain.1 | grep mem
Error: memory_dynamic_max must be less than or equal to
memory_static_max
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
kfraser@localhost.localdomain [Fri, 8 Jun 2007 10:19:55 +0000 (11:19 +0100)]
hvm: Respect irqbase set by protected mode in mode switching with VMXAssist.
RHEL4U4 PAE SMP guest currently crashes, and we found changeset 15214
introduced it. This patch fixes it.
Signed-off-by: Xin Li <xin.b.li@intel.com>
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 19:02:27 +0000 (20:02 +0100)]
x86: Print source of e820 memory map during boot. Fix Xen-e801 memmap
parsing. Get rid of unneeded e820_raw variable -- map straight onto
boot-trampoline e820 array.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 18:41:59 +0000 (19:41 +0100)]
x86: Return to real-mode when booting primary CPU, and gather
memory-map information (from int15{e820,e801,88} and int12 bios
calls).
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 15:44:04 +0000 (16:44 +0100)]
Fix the README to match Xen 3.1
From: Ryan Riley <rileyrd@gmail.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 13:30:21 +0000 (14:30 +0100)]
Add backtrace support to xenoprof.
Signed-off-by: Amitabha Roy <amitabha.roy@gmail.com>
Reviewed-by: Jose Renato G Santos <joserenato.santos@hp.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 10:15:24 +0000 (11:15 +0100)]
ioemu: Tiny indentation fix.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 10:14:18 +0000 (11:14 +0100)]
xentop: Display blktap information.
Signed-off-by: Satoshi UCHIDA <s-uchida@ap.jp.nec.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 10:13:41 +0000 (11:13 +0100)]
xenstat: Adding to scan blktap information.
Signed-off-by: Satoshi UCHIDA <s-uchida@ap.jp.nec.com>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 10:06:00 +0000 (11:06 +0100)]
Fix the vnet module for Xen 3.1.x.
Signed-off-by: Robert Valentan <R.Valentan@solid-soft.at>
kfraser@localhost.localdomain [Thu, 7 Jun 2007 10:02:23 +0000 (11:02 +0100)]
tools: Fix some type issues GCC 4.1.0 warnings.
FC5's gcc 4.1.0 can't make some files in tools/ due to its stronger
type checking.
From: Dexuan Cui <dexuan.cui@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 6 Jun 2007 19:25:58 +0000 (20:25 +0100)]
qemu: Fix vnc port offset in xenstore
The new version of qemu uses 5900 instead of 0 as the port.
There is no need to offset it by 5900 when writing out the
console/vnc-port entry.
Signed-off-by: Ben Guthro <bguthro@virtualiron.com>
kfraser@localhost.localdomain [Wed, 6 Jun 2007 19:24:06 +0000 (20:24 +0100)]
[HVM] Prevent usb driver crashes in Windows
Use atomic updates to read/write usb controller data.
This can be done because:
a) word copies on x86 are atomic
b) The USB spec requires word alignment
This will need to be enhanced once USB 1.2 is supported.
Signed-off-by: Steve Ofsthun <sofsthun@virtualiron.com>
Update to copy 'longword'-sized atoms.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 6 Jun 2007 16:49:39 +0000 (17:49 +0100)]
pv-on-hvm: Allow Linux and Xen paths to be overridden by the environment.
Signed-off-by: Ben Guthro <bguthro@virtualiron.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 6 Jun 2007 16:32:58 +0000 (17:32 +0100)]
Fix pv-on-hvm mkbuildtree script for new repo layout.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Tue, 5 Jun 2007 16:05:13 +0000 (17:05 +0100)]
Synchronise xend-config file comments w.r.t. networking.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Tue, 5 Jun 2007 14:52:46 +0000 (15:52 +0100)]
[PATCH] Check for non-existent local repo
Check for non-existent local clone in select-repository.
Also change the #!/bin/bash to #!/bin/sh since the Makefile is calling
it explicitly with sh, and fix a couple other typos.
Signed-off-by: Aron Griffis <aron@hp.com>
kfraser@localhost.localdomain [Tue, 5 Jun 2007 14:51:59 +0000 (15:51 +0100)]
[ACM] Python-to-C Interface for new ACM hypercalls.
Provide an interface for using recently-added Xen functionality from
python.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
kfraser@localhost.localdomain [Tue, 5 Jun 2007 09:28:59 +0000 (10:28 +0100)]
xen: Boot on PAE systems with no EFER MSR.
Bug tracked down by Stefan Berger.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Tue, 5 Jun 2007 09:08:09 +0000 (10:08 +0100)]
hvm vmx: Clean up vmx_io_instruction().
vmx_io_instruction() is too long, so it's hard to understand or change
something. The attached patch splits it into several readable functions,
making it easer to understand and extend.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
kfraser@localhost.localdomain [Tue, 5 Jun 2007 09:00:15 +0000 (10:00 +0100)]
console daemon: Fix realloc() usage.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Tue, 5 Jun 2007 08:59:34 +0000 (09:59 +0100)]
qemu: Use domid in qemu log name. Can be related back to domain name
or uuid via xend logs, less unique than pids to prevent excessive
number of log files hanging around, yet not immediately reused (unlike
e.g., after domain crash or reboot) where previous log file is
immediately overwritten. Not using domain names in logs avoids and
chance of unprintable charcatres becoming embedded in filenames.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Tue, 5 Jun 2007 07:57:53 +0000 (08:57 +0100)]
BUILD: Clone the Linux tree if it wasn't found via $(LINUX_SRC_PATH)
even if it happens to be a local directory.
If we located the Linux tree by deriving the path from "hg path
default" and it is a local path then prepend "file://" to prevent us
trying to symlink it.
Unfortunately mercurial doesn't understand file:// URLs so we have to
strip it off again before calling hg clone.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Tim Deegan [Mon, 4 Jun 2007 16:09:12 +0000 (17:09 +0100)]
[HVM] Save/restore: trigger FPU state save based on v->fpu_initialised
and not on the i387_valid flag, which has no meaning in this context.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Steven Hand [Mon, 4 Jun 2007 15:47:48 +0000 (16:47 +0100)]
Ensure CR0.TS is set for VMX domains regardless of the current
state of the real TS bit. This fixes a number of subtle FP
corruption issues within PV domains when running alongside VMX ones.
Signed-off-by: Steven Hand <steven@xensource.com>
Ian Campbell [Mon, 4 Jun 2007 14:41:32 +0000 (15:41 +0100)]
If we find a Linux repository in $(LINUX_SRC_PATH) then symlink it
instead of cloning it. This enables developers to keep a linux tree
separate to their xen repository and to make changes there which are
picked up by the Xen build.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 14:32:11 +0000 (15:32 +0100)]
Fix boot loader hangs with syslinux's 32-bit vesamenu module.
Syslinux can load 32-bit UI code for menu handling. But the core of
syslinux is still 16-bit. When it jumps to this 32-bit code, it
installs a set of 32-bit interrupt trap handlers which just bounce the
interrupts back to 16-bit mode.
But this plays badly with vmxassist. When running 16-bit boot loader
code, vmxassist installs its own trap handlers which bounce vPIC
interrupts back down to 16-bit mode. The trap handlers live at
int 0x20 to 0x2f, so when the 16-bit code tries to reprogram the vPIC,
vm86 rewrites the outb()s on the fly to set the irq_base vectors
accordingly.
So when syslinux enters 32-bit mode, the vPIC has still been
programmed to point to vmxassist's bounce traps, even though vmxassist
is no longer active once the guest is running 32-bit code. So the
wrong interrupts get delivered to the guest.
Fix this by restoring the rombios vPIC irq_base vectors when we leave
vmxassist mode, and returning the vmxassist traps when we reenter it.
These irq base values are hard-coded in this patch, but they are
already hard-coded in vmxassist so any boot code that relies on
changing them will already fail.
Signed-off-by: Stephen Tweedie <sct@redhat.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 14:21:12 +0000 (15:21 +0100)]
qemu: More care over filename string sizes.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 14:17:49 +0000 (15:17 +0100)]
Use name instead of pid in QEMU logs.
Change the logfile name generated by qemu-dm to use the pattern
qemu-dm-[NAME].log instead of qemu-dm-[PID].log. This makes it
easier for the adminsitrator to figure out which log corresponds to
which guest (particularly after a crash where you no long know what
PID the qemu-dm process for your guest had). It also prevents the
number of log files from growing unbounded.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 13:49:12 +0000 (14:49 +0100)]
Add persistent guest & hv logging in xenconsoled.
* The --log command line argument takes one of 4 values
- none - no logging (the default)
- hv - log all hypervisor messages
- guest - log all guest messages
- both - log all guest & hypervisor messages
* The --log-dir command line argument takes a path to specify where
to store logfiles. If omitted it defaults to /var/log/xen/console
* The hypervisor logfile is $LOGDIR/hypervisor.log
* The guest logfile is $LOGDIR/guest-[NAME].log
* If receiving a SIGHUP it will close & re-open all log files to
enable logrotate to do its magic
* Fixes the permissions of /var/run/xenconsoled.pid
* Adds a --pid-file command line argument to override the default
location of pid file (this is not really related to logging, but
since I was in that code...)
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 13:40:12 +0000 (14:40 +0100)]
Enable ACM on PPC
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 13:25:03 +0000 (14:25 +0100)]
gdbstub: Bit-field boolean must be unsigned.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 13:24:46 +0000 (14:24 +0100)]
xentrace: Bit-field boolean must be unsigned.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 13:05:10 +0000 (14:05 +0100)]
xen: Fix domain teardown to avoid use-after-free of vcpu structs.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Mon, 4 Jun 2007 11:47:53 +0000 (12:47 +0100)]
Remove a few stray references to the sparse tree.
The only non-obvious change here is in tools/xenfb/Makefile. This now picks up
the headers it requires via the copy in tools/libxc.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Ian Campbell [Mon, 4 Jun 2007 10:37:50 +0000 (11:37 +0100)]
Remove $(LINUX_SRCDIR) on distclean.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Ian Campbell [Mon, 4 Jun 2007 10:16:19 +0000 (11:16 +0100)]
Remove sparse tree.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Mon, 4 Jun 2007 10:01:46 +0000 (11:01 +0100)]
[NET] Remove netloop from network bridge init scripts.
The scripts are partly based on work by Daniel P. Berrange.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Ian Campbell [Mon, 4 Jun 2007 09:14:53 +0000 (10:14 +0100)]
Switch over to separate Linux tree.
The sparse tree is now deprecated and will deleted in due course. The
xen-unstable build system will automatically clone the correct Linux
tree for you and all the old makefile targets
(dist/linux-2.6-xen{,0,U}/etc) should continue to work.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Ian Campbell [Mon, 4 Jun 2007 09:14:23 +0000 (10:14 +0100)]
Added tag sparse-tree-deprecated for changeset
bd3d6b4c52ec809f080c89c4ffcf61dc6e445978
Tim Deegan [Fri, 1 Jun 2007 13:50:52 +0000 (14:50 +0100)]
[XEN] Shadow: cache gva->gfn+rights translations between guest TLB flushes
to make lookups faster for emulation and hvm_copy.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Tim Deegan [Fri, 1 Jun 2007 13:32:11 +0000 (14:32 +0100)]
[XEN] Shadow: emulate a few extra instructions on PAE pagetable writes
in the hope of catching the "other half" write without another enter/exit.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Thu, 31 May 2007 15:09:11 +0000 (16:09 +0100)]
VMX: Only write top-half of machine addresses into VMCS fields in PAE mode.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 31 May 2007 15:05:27 +0000 (16:05 +0100)]
hvm: Pause domain during state save/load.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Thu, 31 May 2007 13:51:09 +0000 (14:51 +0100)]
Updates for XEN_LINUX_SOURCE=hg-clone.
By default search for mercurial repositories in $(LINUX_SRC_PATH)
followed by a location derived from the current Xen repository's
default parent. e.g. A clone of
http://xenbits.xensource.com/xen-unstable.hg will derive
http://xenbits.xensource.com/linux-2.6.18-xen.hg. A specific
repository can be specified by giving XEN_LINUX_HGREPO=<URL or path>
on the make command line.
Add capability to put XEN_LINUX_UPDATE=y on the make command line to
update the Linux repository. Disabled by default since we don't want
to touch a developers repository without permission.
Dropped XEN_LINUX_HGUPDATES since it was unecessary and confusing.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Thu, 31 May 2007 09:58:22 +0000 (10:58 +0100)]
hvm: Exceed maximum number of ioemu's NIC for VNIF.
QEMU should ignore excess NICs instead of failing to initialise.
Signed-off-by: Takanori Kasai <kasai.takanori@jp.fujitsu.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
kfraser@localhost.localdomain [Thu, 31 May 2007 09:53:11 +0000 (10:53 +0100)]
hvm: Allow GNTTABOP_query_size.
Signed-off-by: Takanori Kasai <kasai.takanori@jp.fujitsu.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 31 May 2007 09:42:34 +0000 (10:42 +0100)]
Fix stale email addresses in xenstat and xentop.
Our @us.ibm.com emails cannot forward to us; we do not have access to
them when not interning at IBM. People have attempted to reach us via
those addresses and failed; occasionally they've Googled for a current
address, which made us realize the problem. Change each email to a
more permanent address, so people can reach us in the future. Change
bug reporting addresses to a xen mailing list rather than an
individual.
Signed-off-by: Josh Triplett <josh@kernel.org>
kfraser@localhost.localdomain [Thu, 31 May 2007 09:30:15 +0000 (10:30 +0100)]
Fix the very-slow-IDE-IO-speed issue of Qemu 0.90
c/s 15021 updated Qemu to 0.90. In the new Qemu's ide.c, asynchronous
IO (AIO) is used to replace the previous DMA thread; here when an AIO
request is completed, dom0 sends a signal SIGUSR2 to Qemu (see
block-raw.c: qemu_aio_init(),raw_aio_setup()), then the signal
interrupts the select() in main_loop_wait() at once, next,
qemu_aio_poll() is called to reap a completed AIO request.
However, in certain Linux distributions (i.e., x86_64 RHEL 4u4), if
Qemu is spawned by the python script image.py of Control Panel, the
SIGUSR2 of Qemu is blocked by default due to some reasons (i.e. it
seems Python 2.3.4 has this issue; 2.4.2 is ok); so in most cases the
select() in main_loop_wait() can only time out in 10ms, then the
qemu_aio_poll() is called - this results in a very slow disk IO
speed...
This patch ensures aio_sig_num is unblocked when the AIO is used in
Qemu 0.90.
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
kfraser@localhost.localdomain [Thu, 31 May 2007 08:49:53 +0000 (09:49 +0100)]
hvm: Correctly hook VLAPIC MSR change notification for VMX into
save/restore path.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 17:29:21 +0000 (18:29 +0100)]
Remove 'hpet_force' option. Replace with 'clocksource' option to allow
forced selection of platform clocksource, overriding the default
priority ordering.
Usage: clocksource={pit,hpet,cyclone,acpi}
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 17:12:02 +0000 (18:12 +0100)]
Add support for ACPI PM Timer as platform clock source.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 16:47:00 +0000 (17:47 +0100)]
Clean up around domain init/destroy.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Wed, 30 May 2007 16:37:58 +0000 (17:37 +0100)]
Make header mismatch checking more robust against build time changes in
xen/include/public/foreign.
Unfortunately diff's -x and -X options do not support excluding paths just
individual files so we have to list unqualified filenames. This is ok for now
since there are no filename clashes.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 16:01:26 +0000 (17:01 +0100)]
Simplify APIC_ACCESS VMX support.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 15:48:28 +0000 (16:48 +0100)]
Add VMX memory-mapped Local APIC access optimization.
Some operating systems access the local APIC TPR very frequently, and
we handle that using software-based local APIC virtualization in Xen
today. Such virtualization incurs a number of VM exits from the
memory-access instructions against the APIC page in the guest.
The attached patch enables the TPR shadow feature that provides APIC
TPR virtualization in hardware. Our tests indicate it can
significantly boost the performance of such guests including 32-bit
Windows XP/2003.
Moreover, with the patch, local APIC accesses other than TPR in guests
are intercepted directly as APIC_ACCESS VM exits rather than
PAGE_FAULT VM exits; this can lower the emulation cost of such
accesses.
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
Ian Campbell [Wed, 30 May 2007 12:10:20 +0000 (13:10 +0100)]
Check for mismatch between headers shared by Linux and Xen.
This is only enabled for the linux-2.6-xen tree and is intended to
ensure that the separate Linux tree stays up to date once we
switch away from the sparse tree.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Ian Campbell [Wed, 30 May 2007 10:55:38 +0000 (11:55 +0100)]
Fix preemption of multicalls in compat mode.
mcs->call (struct multicall_entry) always needs to be translated into
mcs->compat_call (struct compat_multicall_entry) when a multicall is
preempted in compat mode. Previously this translation only occured for
those hypercalls which explicitly called hypercall_xlat_continuation()
which doesn't cover all hypercalls which could potentially be
preempted.
Change hypercall_xlat_continuation() to only translate only the
hypercall arguments themselves and not the multicall_entry
layout. Translate the layout for all hypercalls in in
compat_multicall() instead.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 09:47:05 +0000 (10:47 +0100)]
[NET] back: Add lazy copying
This patch adds lazy copying using the new unmap_and_replace grant
table operation.
We keep a list of pending entries sorted by arrival order. We'll
process this list every time net_tx_action is invoked. We ensure
that net_tx_action is invoked within one second of the arrival of
the first packet in the list.
When we process the list any entry that has been around for more
than half a second is copied. This allows up to free the grant
table entry and return it to domU.
If the new grant table operation is not available (e.g., old HV
or architectures that don't support it yet) we simply copy each
packet as we receive them using skb_linearize. We also disable
SG/TSO if this is the case.
By default the new code is disabled. In order to enable it,
the module needs to be loaded with the argument copy_skb=1.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
kfraser@localhost.localdomain [Wed, 30 May 2007 09:46:13 +0000 (10:46 +0100)]
[LINUX] gnttab: Add basic DMA tracking
This patch adds basic tracking of outstanding DMA requests on
grant table entries marked as PageForeign.
When a PageForeign struct page is about to be mapped for DMA,
we set its map count to 1 (or zero in actual value). This is
then checked for when we need to free a grant table entry early
to ensure that we don't free an entry that's currently used for
DMA.
So any entry that has been marked for DMA will not be freed early.
If the unmapping API had a struct page (which exists for the sg
case) then we could do this properly.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
kfraser@localhost.localdomain [Wed, 30 May 2007 09:45:44 +0000 (10:45 +0100)]
[XEN] gnttab: Add new op unmap_and_replace
The operation unmap_and_replace is an extension of unmap_grant_ref.
A new argument in the form of a virtual address (for PV) is given.
Instead of modifying the PTE for the mapped grant table entry to
null, we change it to the PTE for the new address. In turn we
point the new address to null.
As it stands grant table entries once mapped cannot be
remapped by the guest OS (it can however perform a new
mapping on the same entry but that is within our control).
Therefore it's safe to manipulate the mapped PTE entry to
redirect it to a normal page where we've copied the contents.
It's intended to be used as follows:
1) map_grant_ref to v1
2) ...
3) alloc page at v2
4) copy the page at v1 to v2
5) unmap_and_replace v1 with v2
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Added compat integration (PAE-on-64).
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 09:06:23 +0000 (10:06 +0100)]
xend: Don't destroy domains on shutdown timeout.
Do not destroy domains on shutdown timeout: the administrator needs to
be able to decide what to do if a domain does not shut down.
Signed-off-by: John Levon <john.levon@sun.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 09:04:23 +0000 (10:04 +0100)]
Fix xenstore entry accounting
xenstored is incorrectly accounting domain nodes when transactions
fail. Store pending count changes in the transaction structure, and
apply at transaction completion, instead of directly applying the
changes.
Signed-off-by: Max Zhen <max.zhen@sun.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 08:59:09 +0000 (09:59 +0100)]
Fix xen build on NetBSD.
From: Christoph Egger <Christoph.Egger@amd.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 30 May 2007 08:51:38 +0000 (09:51 +0100)]
xen: Remove assumption that first NUMA node to be discovered is node0.
Based on a patch by Alex Williamson.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 16:26:26 +0000 (17:26 +0100)]
pv-on-hvm: Re-init hypercall stubs page after HVM save/restore.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 15:06:36 +0000 (16:06 +0100)]
x86: Sanitise the MFN passed to mod_l1_entry().
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Fri, 25 May 2007 14:28:52 +0000 (15:28 +0100)]
Correct 32on64 handling of VCPUOP_register_runstate_memory_area. We
were copying too many bytes from the guest so the test for
area.addr.h.c != area.addr.p was failing.
Added a WARN_ON to the kernel to catch this case. It would be a BUG_ON
but this would break the new kernel on older hypervisors and the only
real problem is that stolen time is not updated, which we can live
with.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 12:41:01 +0000 (13:41 +0100)]
kfraser@localhost.localdomain [Fri, 25 May 2007 12:02:08 +0000 (13:02 +0100)]
xend: Fix PAM authentication method for some versions of PyPAM.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Fri, 25 May 2007 10:24:02 +0000 (11:24 +0100)]
LINUX/x86_64: Cleanup 15129:
a40967e39652 by using __pgd rather than
abusing mk_kernel_pgd.
Also set the user bit on the vsyscall entry in the user pgd.
Both changes suggested by Jan Beulich.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Ian Campbell [Fri, 25 May 2007 10:03:40 +0000 (11:03 +0100)]
Factor out sparse tree makefile fragments and introduce hg-clone and
tarball methods in preparation for migrating away from the sparse tree
layout.
This re-introduces 14883:
6568c1882af5 and 14862:
98efd2e410ae plus
further fixes.
Also add makefile fragments to support building the paravirt ops tree
(make linux-2.6-paravirt-build).
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 09:53:07 +0000 (10:53 +0100)]
xend: Fix checkname so that it detects duplicate domains.
Signed-off-by: Mats Petersson <mats.petersson@amd.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 09:23:16 +0000 (10:23 +0100)]
[XEND/XM] Improve error reporting in 'xm restore'.
From: Mats Petersson <Mats.Petersson@amd.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 09:17:43 +0000 (10:17 +0100)]
linux: Fix PV-on-HVM crash when installing PV drivers on native system.
From: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 09:16:00 +0000 (10:16 +0100)]
xend: Increase shutdown timeout.
30 seconds is not a very long time for a busy domain to cleanly
shutdown all its services, which may be arbitrary in
complexity. Increase the timeout to 5 minutes.
Signed-off-by: John Levon <john.levon@sun.com>
kfraser@localhost.localdomain [Fri, 25 May 2007 09:15:37 +0000 (10:15 +0100)]
Do not destroy domains that timeout when shutting down.
Instead of violently destroying a domain that is not responding to a
shutdown request, rename the domain to indicate the problem and leave
it alone; this allows the admin to make corrective actions (which may
or may not include destroying the domain).
Signed-off-by: John Levon <john.levon@sun.com>
kfraser@localhost.localdomain [Thu, 24 May 2007 14:45:43 +0000 (15:45 +0100)]
xm: Extend width of ID in 'xm list' to 5 digits.
The Domain ID is currently expected to fit in 3 digits.
When doing extensive testing on SAVE/RESTORE I can easily get ID's
into the 4-digit range, which makes some of the formatting to get out
of sync. Extend the width to 5 digits (and change some %-ns to %ns, as
the ID for example is right justified, %-ns is left-justified, which
didn't look too good).
Signed-off-by: Mats Petersson <mats.petersson@amd.com>
kfraser@localhost.localdomain [Thu, 24 May 2007 14:40:54 +0000 (15:40 +0100)]
xm-test: ia64 min memory & arbitrarily large ramdisk
1) Sets a minimum of 128MB RAM for an ia64 domU; this limit was set
after experimentation, which seems to indicate it's a reasonable lower
limit (the 32MB limit previously in place did not allow an ia64 domU
to start). If there's any problem with this, I don't mind splitting
the patch and sending it to the ia64 list, but it was small, so I hope
it's okay to include it.
2) xm-test uses ramdisks built with uClibc, which doesn't compile on
ia64. I was able to create a ramdisk by hand, but as it was too
large, the resultant domU crashed after boot. This patch enables the
use of an arbitrarily large ramdisk, so long as it's uncompressed. As
xm-test builds only uncompressed ramdisks, I figured this was
reasonable for that specific application. Suggestions on how to
detect and handle a compressed ramdisk would be welcome.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
kfraser@localhost.localdomain [Thu, 24 May 2007 14:36:13 +0000 (15:36 +0100)]
xend: Give "xm start" a -c parameter for connecting to the console
From: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 24 May 2007 14:30:05 +0000 (15:30 +0100)]
xend: Add site-packages into Python search path
Signed-off-by: S.Caglar Onur <caglar@pardus.org.tr>