Skipping image name at start of command line is an
architecture-specific action. Definitely not required on IA64, for now
make it just x86 specific. Signed-off-by: Keir Fraser <keir@xensource.com>
hvm: Intercept reads of MSR_IA32_MCG_CAP and return 0.
This indicates no machine check "units", which agrees more closely
with Xen's super-minimal machine check architecture (just enough to
allow Windows to run). This fixes a bug that occurs when migrating a
RHEL4-64bit guest to a host with fewer machine check units than the
original host. These host physical details shouldn't be leaking
through to guests.
Signed-off-by: David Lively <dlively@virtualiron.com>
x86: Remove (most) Centaur CPU support. Only VIA C7 can work, as it
has CMOV support. Leave a small amount of centaur.c around to support
that. MTRR code goes entirely, as 686-class Centaur CPUs have generic
MTRR support.
Don't carry dead code needlessly: this is a family 5 CPU, which Xen
doesn't support. Perhaps, other CPUs' files could use some cleanup in
that respect, too, but there it would increase the delta to the Linux
origin of these files, while here the entire file can go away.
ioemu: Avoid struct members clashing with POSIX apis
The TPM code in tools/ioemu/hw/tpm_tis.c has a struct containing a
number of function pointers with names open, close, read, write which
are the same as various POSIX apis already #included in the
file. POSIX allows these functions to be defined as macros and latest
GCC/glibc does indeed define them as macros depending on compiler
flags. This causes compile errors when deferencing the struct
members. The solution is either to change calls like ctx->open () to
be (* ctx->open) (), or simply to rename the struct members. Since
this struct was only used inside that one file I simply renamed them.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
When I tested xm create command, I saw the following error message.
I expected an error message "Error: (12, 'Cannot allocate memory')"
because I intentionally caused a memory shortage on the test.
But the error message was different from my expectation.
# xm create /xen/HVMdomain.1
Using config file "/xen/HVMdomain.1".
Error: an integer is required
I looked at xend.log to examine the cause why the error message was
shown. (Could you see the attached xend.log?)
xend had the error message "Error: (12, 'Cannot allocate memory')"
first. But xend changed the error message to "Error: an integer is
required" halfway. I'm not sure about the cause why an exception
occurred in logging processing. But when I applied an attached patch,
I confirmed that the error message that I expected was shown. The
patch does not call xc.domain_destroy_hook() if self.domid is None.
ioemu: error checkin when setting up the Cirrus Logic video device.
set_mm_mapping() may fail because of xc_domain_populate_physmap(). In
this case, we should not blindly go on; the xc_map_foreign_batch()
that follows will cause a page fault and, at best, get mapped in a
zeroed page from the dom0 (which is not what we want). While I'm in
here, fix a memory leak on an error path.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Fix VNC server after HVM save/restore/migrate, when no vncpasswd specified.
The basic problem was that the "image" section of the sxp had a "None"
in it, which means that on the restore (either on the local machine or
remote machine), the vnc server in the qemu device model literally
needed the string "None" to properly connect. This simple patch only
puts a vncpasswd entry in the image if it is *not* None in the python
code, thus avoiding the whole issue.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
[xen, xencomm] xencomm multiple page support
Current implementation doesn't allow struct xencomm_desc::address
array to be more than single page. On IA64 it causes 64GB+ domain
creation failure. This patch generalizes xencomm to allow multipage
[xen, xencomm] fix various xencomm invalid racy access.
- Xencomm should check struct xencomm_desc alignment.
- Xencomm should check whether struct xencomm_desc itself (8 bytes)
doesn't cross page boundary. Otherwise a hostile guest kernel can
pass such a pointer that may across page boundary. Then xencomm
accesses an unrelated page.
- Xencomm shouldn't access struct xencomm_desc::nr_addrs multiple
times. Copy it to local area and use the copy.
Otherwise a hostile guest can modify at the same time.
- Xencomm should check whether struct xencomm_desc::address[] array
crosses page boundary. Otherwise xencomm may access unrelated pages.
- Xencomm should get_page()/put_page() after address conversion from
paddr to maddr because xen supports SMP and balloon driver.
Otherwise another vcpu may free the page at the same time.
Such a domain behaviour doesn't make sense, however nothing prevents
it.
Alex Williamson [Thu, 16 Aug 2007 15:37:54 +0000 (09:37 -0600)]
[IA64] Shrink vtlb size
Instrumenting thash_purge_all() shows a very low usage of vtlb
entries (21 at most).
This patch shrinks the default vtlb size from 512KB to 16KB
to optimize memory. This also speeds up ptc_e emulation.
To improve the hash function, frequency of collision never changed
and there is no performance degradation.
[xen, xencomm] preparetion for xencomm consolidation.
Xen/powerpc runs in real mode so that it uses maddr interchangably
with vaddr. But it isn't the case in xen/ia64. It is necessary to
convert maddr to vaddr to access the page. maddr_to_virt() doesn't
convert on powerpc, so it should work on both archtechture. Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
xen: Do not accept DOMID_SELF as input to DOMCTL_getdomaininfo.
This was screwing up callers that loop on getdomaininfo(), if there
was a domain with domid DOMID_FIRST_RESERVED-1 (== DOMID_SELF-1).
They would see DOMID_SELF-1, then look up DOMID_SELF, which has domid
0 of course, and then start their domain-finding loop all over again!
Found by Kouya Shimura <kouya@jp.fujitsu.com>. Thanks! Signed-off-by: Keir Fraser <keir@xensource.com>
This patch adds missing includes that currently work through indirect
inclusions. This cannot be relied on and indeed does break on older
kernels (2.4 for PV-on-HVM).
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
- To remove device info, it waits for the backend path of the device
to be removed.
- It removes device info from domain info.
- It saves domain info to the config.sxp of the managed domain.
- prevent vlans from being relabeled when they are in use by the
current policy
- fix problems when doing an update of the policy and the name of the
policy changes while doing that
- refactor code that has to take into consideration that unlabeled
domains may be defined using the label __UNLABELED__
- make 'xm list --label' show the complete label of a domain
Clean up usage of 'current' in do_iret() hypercall. Signed-off-by: Christoph Egger <Christoph.Egger@amd.com> Signed-off-by: Keir Fraser <keir@xensource.com>
Alex Williamson [Thu, 9 Aug 2007 02:48:11 +0000 (20:48 -0600)]
[IA64] Saner dom0 memory and cpu defaults
Some ia64 xen dom0 tweaks:
* Increase default memory allocation from 512M to 4G
* Increase default vcpu allocation from 1 to 4
* Implement rough calculation of what the maximum memory
that can be safely allocated to dom0 is
* If need be, scale down requested memory allocation to fit
available memory, rather than simply panicking
* If dom0_mem=0 is specified, allocate all available mem
hvm: Handle hw_cr[] array a bit more sanely.
SVM for the most part does not need to use it at all, and this makes
the code clearer. Signed-off-by: Keir Fraser <keir@xensource.com>
xentop: fix abnormal value cpu(%) when domain shutdown.
If we test multiple domain create/shutdown many times.
sometimes abnormal values of cpu(%) are appeared.
This is because shutdown domain remove failure.
(in other words, memory corruption of struct )
This corruption makes abnormal cpu(%) values are shown xentop
sometimes.
[ACM] Explicitly cast byte arrays when used in copy_from_guest
Explicitly cast the target array type to an array of 'u8' since the
number of bytes copied is calculated from the type of element in the
target array times the number of elements (3rd parameter).
xm: Fix control characters causing xm tests to fail.
On my FC7 installation some of the test in the xm test suite fail for
no apparent reason. The problem is that the python 'readline' module
puts out control characters that confuse code that parses the output
of some of the xm commands that are being run. The readline module
dumps those characters upon import. The attached patch works around
this problem.
Here's what I get in out.txt when doing 'python 2>&1 | tee out.txt'
and 'import readline' on the python command line:
Python 2.5 (r25:51908, Apr 10 2007, 10:29:13)
[GCC 4.1.2 20070403 (Red Hat 4.1.2-8)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
[[?1034h>>> >>>
The characters before the '>>>' above are those printed by that
version of 'readline' and remain invisible on the screen but appear in
the file and confuse the xm test suite parser.
hvm: Fix save/restore when callback_via line is routed through a PCI
INTx wire. Signed-off-by: Edwin Zhai <edwin.zhai@intel.com> Signed-off-by: Keir Fraser <keir@xensource.com>
When calling destroyDevice code path (e.g. xm block-detach dom devid),
allow specifying an integer device id or a device name such as xvdN or
/dev/xvdN. Allowing the /dev/xvdN form is useful when detaching
devices from dom0. Bootloaders may do this to unmount a disk
previously mounted in dom0.
Move examination of device ID format into the DevController,
permitting device controllers to determine a valid device ID instead
of higher level code.
Add support for removing/replacing a policy from the running
system. This operation is only successful if currently running VMs
would also be allowed to run under the new policy. Removing the
current policy means that the default policy is installed, which then
only has support for a single VM label and STE type (SystemManagement)
and is the same policy that the system starts up with when no policy
is chosen.
[ACM] Support for running unlabeled domains alongside labeled ones
Add support for running unlabeled domains alongside labeled ones, if
the policy contains a VM label with name '__UNLABELED__' and an STE
type with the same name. The ezpolicy tool has been modified to
automatically suggest a policy under which unlabeled domains can
run. The user may delete this, if this is not desired.
On debug builds, scrub pages with non-zero poison.
Will flush out guests which are relying on zeroed memory. Signed-off-by: Keir Fraser <keir@xensource.com>
libelf: Clean up and fix up BSD symtab handling.
Should now (hopefully) work for domU and dom0 builders. Signed-off-by: Keir Fraser <keir@xensource.com>
projects / xen.git / log