Keir Fraser [Thu, 18 Jun 2009 09:14:16 +0000 (10:14 +0100)]
x86_64: allow more vCPU-s per guest
Since the shared info layout is fixed, guests are required to use
VCPUOP_register_vcpu_info prior to booting any vCPU beyond the
traditional limit of 32.
MAX_VIRT_CPUS, being an implemetation detail of the hypervisor, is no
longer being exposed in the public headers.
The tools changes are clearly incomplete (and done only so things
would
build again), and the current state of the tools (using scalar
variables all over the place to represent vCPU bitmaps) very likely
doesn't permit booting DomU-s with more than the traditional number of
vCPU-s. Testing of the extended functionality was done with Dom0 (96
vCPU-s, as well as 128 vCPU-s out of which the kernel elected - by way
of a simple kernel side patch - to use only some, resulting in a
sparse
bitmap).
ia64 changes only to make things build, and build-tested only (and the
tools part only as far as the build would go without encountering
unrelated problems in the blktap code).
Keir Fraser [Wed, 17 Jun 2009 06:39:27 +0000 (07:39 +0100)]
xend: pass-through: Clean up hvm_destroyPCIDevice()
There seems to be little need to use the domain, bus, slot and
function to look up the virtual slot to pass as the argument to
hvm_destroyPCIDevice(), only to have hvm_destroyPCIDevice() use the
virtual slot for the sole purpose of looking up the domain, bus, slot
and function.
Keir Fraser [Wed, 17 Jun 2009 06:36:47 +0000 (07:36 +0100)]
xend: pass-through: Move pci conversion functions to pci.py
Move dev_dict_to_sxp() into XendSXPDev.py. Move
pci_convert_dict_to_sxp() and pci_convert_sxp_to_dict() to pci.py,
where other similar functions live. This makes these functions
accessible outside of the XendConfig class.
Keir Fraser [Wed, 17 Jun 2009 06:34:59 +0000 (07:34 +0100)]
xend: pass-through: Common parse_pci_name()
Share some parsing code between different parts of xm.
This has the side-effect that the device specification for
hot-plug may now include the VSLOT and OPTS as per device
specifictions in the domain configuration file.
Keir Fraser [Wed, 17 Jun 2009 06:21:03 +0000 (07:21 +0100)]
Introduce mechanism to check standard conformance of headers
While pretty simplistic, it appears to serve the purpose at the moment
(i.e. it spotted two places where a GNU extension was used withou
proper preprocessor conditionals). The "simplistic" here includes that
the checking gets only done for native builds, and ia64 gets excluded
due to its arch-specific header intentionally (for whatever reason)
checking that anonymous struct/unions can be used.
Keir Fraser [Wed, 17 Jun 2009 06:14:02 +0000 (07:14 +0100)]
tmem: fix 32-on-64 support
This implicitly required coverting the tmem_op structure from
anonymous to standard struct/union sub-fields, and extending the
get-fields.sh helper script to deal with typedef-ed guest handles used
as types of translated compound type fields.
Keir Fraser [Tue, 16 Jun 2009 12:57:18 +0000 (13:57 +0100)]
x86: improve output resulting from sending '0' over serial
While the original logic already implied that the kernel part of the
guest's address space is identical on all vCPU-s (i.e. for all guest
processes), it didn't fully leverage the potential here: As long as
the top page table currently active is owned by the subject domain
(currently only Dom0), the stack dump can be done without extra
effort.
For x86-64, additionally add page table traversal so that the stack
can be dumped in all cases (unless it's invalid or user space).
I left the 32-bit variant of do_page_walk() unimplemented for the
moment as I couldn't convince myself using map_domain_page() there is
a good idea, and didn't want to introduce new fixmap entries either.
Keir Fraser [Tue, 16 Jun 2009 12:52:13 +0000 (13:52 +0100)]
x86/hvm: don't pass through port 0x80 in a few special cases
In a recent commit (99f85a28a78e96d28907fe036e1671a218fee597), KVM
disabled the passthrough of this port due to known problems on certain
HP laptops (see
http://lkml.indiana.edu/hypermail/linux/kernel/0712.3/0872.html
and http://lkml.indiana.edu/hypermail/linux/kernel/0801.0/2388.html).
For Xen, don't do this globally, but rather based on a DMI black list.
Keir Fraser [Tue, 16 Jun 2009 12:41:17 +0000 (13:41 +0100)]
vtd: ats and queued invalidation cleanup
Use iommu_qinval in place of qinval_enabled flag. Use
ecap_queued_inval() for determining whether queued invalidation is
available on this vt-d engine or not.
Keir Fraser [Tue, 16 Jun 2009 10:47:09 +0000 (11:47 +0100)]
tmem: cleanups
- don't mis-use guest handle for passing an MFN value
- eliminate unnecessary (and misplaced) use of XEN_GUEST_HANDLE_64
- use copy_from_guest() instead of __copy_from_guest() for loading the
argument structure
Keir Fraser [Tue, 16 Jun 2009 10:31:20 +0000 (11:31 +0100)]
vtd: Clean up lock for VT-d register writes
It should get lock to write VT-d registers. Currently there are some
register writes without lock. This patch complements register_lock for
those writes.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Keir Fraser [Tue, 16 Jun 2009 10:30:45 +0000 (11:30 +0100)]
vtd: Fix flush for SRTP and SIRTP set
SRTP (Set Root Table Pointer) operation must be set before enable or
re-enable DMA remapping. And after set it, software must globally
invalidate the context-cache and then globally invalidate the
IOTLB. This is required to ensure hardware uses only the remapping
structures referenced by the new root-table pointer, and not stale
cached entries. Similarly, SIRTP (Set Interrupt Remap Table Pointer)
operation must be set before enable or re-enable Interrupt
remapping, and after set it, software must globally invalidate the
interrupt entry cache. This patch adds global context and iotlb
flush after set root entry, and globally flushs interrupt entry
cache before enabling Interrupt remapping. And remove the
iommu_flush_all in iommu_resume becuase it becomes redundant after
adds flush for SRTP in init_vtd_hw.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Keir Fraser [Tue, 16 Jun 2009 10:18:32 +0000 (11:18 +0100)]
tmem: fix minor accounting error
Reset a counter when all tmem pages are released. This
only affects status reporting (as displayed by xm tmem-list
or the just patched xenballoon-monitor) but the incorrectly
reported result is misleading.
Signed-off-by: Dan Magenheimer <dan.magenheimer@oracle.com>
Keir Fraser [Tue, 16 Jun 2009 10:17:28 +0000 (11:17 +0100)]
tools/xenballoond: add tmem capability to directed/self-ballooning
and monitor tool
This patch adds tmem support to the largely unknown/unused
xenballoond scripts that implement both self-ballooning and
a foundation for directed-ballooning. Tmem and automated
ballooning are highly complementary in that, when ballooning
is over-aggressive, paging and swapping can increase noticably.
Precache preserves evicted pages that may be needed again soon
(thus eliminating disk reads) and preswap provides memory-based
swapping that occurs if ballooning is insufficiently responsive
to a sudden increase in activity and memory demand (thus
eliminating disk writes and reads).
There are two changes in this patch:
1) The xenballoond service is a convenient place to implement
userland "preswap shrinking".**
2) The xenballoon-monitor script is a convenient place to
report (and view with "watch -d") the frequent memory
rebalancing that results from tmem usage on a busy system.
Note that for best results (and for the monitor script to
work), the xenstore-* tools should be installed on each guest.
(** In a disk-based swap device, stale pages are often left
on-disk even after they are no longer needed or valid; they
are simply overwritten if/when the disk blocks are needed
again, which may be a very long time. Preswap behaves
much like a disk, but uses precious pages of memory that
count against a guest's memory allocation; thus stale
pages are very undesirable. Preswap shrinking periodically
attempts to remove stale pages from preswap by using a
sysfs interface created by the linux-side tmem patch.)
Signed-off-by: Dan Magenheimer <dan.magenheimer@oracle.com>
Keir Fraser [Tue, 16 Jun 2009 10:00:29 +0000 (11:00 +0100)]
xend: allow config file compatibility with new tap syntax
Recently the format of the tap syntax in the config file changed to
using a 4 part specifier (tap:tapdisk:<format>:<filename>) instead
of the old 3-part one (tap:<qcow>:<filename>).
This breaks compatibility with existing config files: a guest start
will throw a Python exception and will be aborted.
AFAICS currently tap:tapdisk is redundant, so the attached patch
simply catches the above mentioned exception and tries to parse the
old format in this case.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Keir Fraser [Tue, 16 Jun 2009 09:58:56 +0000 (10:58 +0100)]
x86 mca: Support MCA recovery actions for latest Intel platforms
When an UC = 1 PCC = 0 MCE happens, there're more types of software
recoverable error defined on on latest Intel Platform. For identifying
those new errors, some new bit (S/AR bit) is defined in MCi_STATUS
register. Also we need MCACOD help's to judge the detailed
error. Combined with the OVER bit, different recovery policies are
required for containing those new errors.
SRAO error is an software recoverable MCA error, no recovery action
required. while SRAR is an software recoverable MCA error, recovery
action is required.
Signed-off-by: Liping Ke <liping.ke@intel.com> Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Keir Fraser [Fri, 5 Jun 2009 08:31:23 +0000 (09:31 +0100)]
blktap2: README updates
As promised, this brings the long out-of-sync documentation up to
date, and adds some getting started information about tapdisk driver
development - I get the occasional email on this latter subject.
Keir Fraser [Fri, 5 Jun 2009 08:29:42 +0000 (09:29 +0100)]
VT-d: correct way to submit command to GCMD register
Per VT-d spec, software should submit only one "incremental" command
at a time to Global Command reigster. Current implementation uses a
variable (gcmd) to record the state of Global Status register. It's
error prone.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Keir Fraser [Fri, 5 Jun 2009 08:27:18 +0000 (09:27 +0100)]
VT-d: define a macro for waiting hardare completion
When set some registers of VT-d, it must wait for hardware
completion. There are lots of duplicated code to do that. This patch
defines a macro for it, thus it is much cleaner.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Keir Fraser [Fri, 5 Jun 2009 08:26:39 +0000 (09:26 +0100)]
VT-d: remove useless variables
This patch removes global variable "vtd_enabled", which is
redundant. "iommu_enabled" is enough. And also removes useless global
variables qi_ctrl and ir_ctrl, which are not used at all.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Keir Fraser [Thu, 4 Jun 2009 21:25:10 +0000 (22:25 +0100)]
x86: hap dirty vram tracking
Currently HAP systems suffer a significant performance loss when a vnc
client is connect or the sdl interface is used, because HAP is lacking
an implementation of track_dirty_vram.
As a consequence qemu always tries to update the whole screen because
it does not know which areas of the screen have been updated by the
guest.
This patch implements track_dirty_vram for HAP enabling the logdirty
mechanism only in a specific gfn range and adding a
paging_log_dirty_range function that returns the log dirty bitmap in a
requested range.
Paging_log_dirty_range is different from paging_log_dirty_op because
operates on a range and also because it does not pause the domain. In
order not to lose any update I moved clean_dirty_bitmap at the
beginning of the function before evaluating the logdirty bitmap.
The bitmap is still safe because it is protected by the logdirty lock.
Keir Fraser [Thu, 4 Jun 2009 09:47:56 +0000 (10:47 +0100)]
xend: pci: only extract the exact pci BDFs
On some hosts:
[root@localhost ~]# ls /sys/bus/pci/devices/0000:00:05.0/
0000:00:05.0:pcie00 0000:05:00.0 class driver local_cpus
resource subsystem_vendor
0000:00:05.0:pcie01 broken_parity_status config enable modalias
subsystem uevent
0000:00:05.0:pcie02 bus device irq power
subsystem_device vendor
Here we should only get 0000:05:00.0, but we also get 0000:00:05.0
unexpectedly. With this patch, xend only extracts the exact BDF(s).
Keir Fraser [Thu, 4 Jun 2009 09:46:13 +0000 (10:46 +0100)]
xm: Don't die when trying to conect the console to short-lived domains
As observed by Mick Joran, if short-lived domain exits cleanly
then os.waitpid() will throw the following exception. This appears
to be because the child process that is used to start the domain
has detached from its parent.
OSError: [Errno 10] No child processes
Cc: Mick Jordan <Mick.Jordan@sun.com> Signed-off-by: Simon Horman <horms@verge.ent.au>
Keir Fraser [Thu, 4 Jun 2009 09:45:24 +0000 (10:45 +0100)]
blktap2: fix parallel Make.
sub make in tools/blktap2/daemon/lib and tools/lvd/lib
can be triggered many times at the same time which results in
weired link error because one target is linking a library while
another target is trying to recreate the library.
Keir Fraser [Thu, 4 Jun 2009 09:41:50 +0000 (10:41 +0100)]
xm: xen-api, pass-through: create: Use vslot for hotplug_slot
Using func for hotplug_slot is not correct, although func is often
zero, previously zero meant please pick a vslot and asking xend to
pick a vslot was the only method available.
This resolves the following error when using Xen API:
$ xm create hvm.conf
...
Internal error: Timed out waiting for device model action.
As the domain, bus, slot and func values are returned as string
representations of decimal, it makes sense for get_hotplug_slot() to
also return string representations of decimal.
As it is, the int() conversion will break cause xm to fail with
an error if the vslot is in the range 0xa-0xf or 0x1a-0x1f.
$ xm pci-list debian
Error: Invalid argument.
And the int() conversion will return the wrong value if
the vslot is in the range 0x10-0x19.
This patch also alters XendDPCI to store hotplug_vslot as an integer
rather than a string. This is consitent with the way other
values are stored inside XendDPCI.
get_hotplug_slot() returning a string is not consistent
with other calls inside XendDPCI, which return integers.
Keir Fraser [Thu, 4 Jun 2009 09:39:32 +0000 (10:39 +0100)]
xend: pass-through: prefix vslot with 0x in device configration
I don't know of the historical reasons for this, but by convention
hex values are stored without a leading '0x' in the backend and
with a leading '0x' in the device configuration.
This patch also removes handling of the case where vslot is missing
from the backend, should never occur.
Keir Fraser [Thu, 4 Jun 2009 09:36:36 +0000 (10:36 +0100)]
libxc: fix link error on ia64
On ia64, xen-unstable 19698:f72d26c00002 cannot be built:
../../tools/libxc/libxenguest.so: undefined reference to
`xc_core_arch_map_p2m_writable'
../../tools/libxc/libxenguest.so: undefined reference to `xc_map_m2p'
Because xc_offline_page.c requires xc_map_m2p() in xc_domain_save.c,
xc_offline_page.c must be compiled only if CONFIG_MIGRATE=3Dy.
Keir Fraser [Wed, 3 Jun 2009 11:59:44 +0000 (12:59 +0100)]
vtd: Fix apic pin to interrupt remapping table index
Originally, it calls xmalloc to set index in
ioapic_rte_to_remap_entry(). When make with debug=y, it may trigger
spinlock BUG_ON because allocate memory with interrupt disabled.
Signed-off-by: Weidong Han <weidong.han@intel.com> Signed-off-by: Keir Fraser <keir.fraser@citrix.com>
Keir Fraser [Wed, 3 Jun 2009 10:11:50 +0000 (11:11 +0100)]
blktap: fix empty QCOW images (bug 1430 part 2)
Empty QCOW images consist of only the L1 table, this results in a
file size which is not sector-aligned. Since blktap uses O_DIRECT, the
block aligned read of the L1 table will go beyond the end of file and
thus returns the actual file size and not the expected length.
This patch checks whether at least the L1 table has been read.
This should fix bug 1430.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Keir Fraser [Wed, 3 Jun 2009 10:11:04 +0000 (11:11 +0100)]
blktap: fix and use ROUNDUP macro (bug 1430 part 1)
As pointed out in Xen Bugzilla 1430 in the blktap QCOW driver the
rounding function is wrong in line 824 of block-qcow.c.
This patch replaces this (and other roundings) with the already
existing ROUNDUP macro (and fixes the usual macro pitfall).
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Keir Fraser [Wed, 3 Jun 2009 10:10:07 +0000 (11:10 +0100)]
blktap2: human readable output for tapdisk2 creation problems
This patch fixes the "file object has no attribute find" failure
we've been seeing when starting blktap2 devices and adds more
meaningful error output to conditions where the tapdisk2 process is
unable to create a blktap2 device.
Keir Fraser [Tue, 2 Jun 2009 10:50:16 +0000 (11:50 +0100)]
xend: pci: improve the assignability checking
1) fix some small typos in util/pci.py;
2) find_all_the_multi_functions(): BDFs of a multi-function PCIe
device could be different in all the 3 fields (bus, device, function),
so we need self.find_parent() and list all t
he BDFs below the parent;
3) to assign a device of the must-be-co-assigned devices, we require
all the related devices should be owned by pciback;
4) detect and disallow duplicate pci string specified in guest config
file due to carelessness.
Keir Fraser [Mon, 1 Jun 2009 17:37:27 +0000 (18:37 +0100)]
tmem: shared ephemeral (SE) pool (clustering) fixes
Tmem can share clean page cache pages for Linux domains
in a virtual cluster (currently only the ocfs2 filesystem
has a patch on the Linux side). So when one domain
"puts" (evicts) a page, any domain in the cluster can
"get" it, thus saving disk reads. This functionality
is already present; these are only bug fixes.
- fix bugs when an SE pool is destroyed
- fixes in parsing tool for xm tmem-list output for SE pools
- incorrect locking in one case for destroying an SE pool
- clearer verbosity for transfer when an SE pool is destroyed
- minor cleanup: merge routines that are mostly duplicate
Signed-off-by: Dan Magenheimer <dan.magenheimer@oracle.com>
Keir Fraser [Mon, 1 Jun 2009 13:15:48 +0000 (14:15 +0100)]
libxc: Exchange a page for PV guest
This patch support exchange a page for a suspended PV guest from user
space.
The basic idea to offline a page is:
1) mark a page offline pending
2) If the page is owned by a HVM domain, user have to live migrate it.
In future, with stub-domain support, we can also exchange the page
without migration.
3) If the page is owned by a PV domain, we will try to exchange the
offline pending page to a new one and free the old page.
This patch achieves item 3.
The method to exchange the offline pending page for PV domain is:
1) Suspend the guest.
2) If the page is being granted out, return with offline pending.
3) Get a copy for the content
4) Scan all page table page to see if any reference to the offending
page, if yes, make the entry to be non-present to reduce the reference
count.
5) After update all page tables, user space tools will try to exchange
the old page. If the new mfn has no reference anymore (i.e.
count_info & count_mask =3D 1), the exchange will allocate a new page,
update the m2p and return success, otherwise it will return fail.
6) If step 5 is success, user space tools will update the content of
the new page change the p2m table, and change all entries scaned in
step 4 to point to new entry.
if step failed, it will try to undo step 4 to revert page table.
7) Resume the guest.
Please refer to thread in
http://www.mailinglistarchive.com/xen-devel@lists.xensource.com/msg63084.html
for more information.
Keir Fraser [Mon, 1 Jun 2009 13:13:53 +0000 (14:13 +0100)]
libxc: Export xc_core_arch_map_p2m_writable()
This patch firstly change the xc_core_arch_map_p2m() to map the p2m to
be writable, then it export this function.
One notice for this patch is, caller should make sure change the p2m
in flight will not cause trouble.
projects / xen.git / log