merge with xen-unstable.hg

[IA64][HVM][QEMU] Add buffer IO mechanism for IA64/VTi domain.

Signed-off-by:Zhang xiantao <xiantao.zhang@intel.com>

[IA64][HVM] Add buffer IO mechanism for Xen/VTi domain. Current
implementation can accelerate Windows guest's dense IO operations
at boot time.

Signed-off-by: Zhang xiantao <xiantao.zhang@intel.com>

[LIBXC] Add -Wmissing-prototypes to CFLAGS, fix warnings resulting from that.

Signed-off-by: Gerd Hoffmann <kraxel@suse.de>

[XEN] Clean up NUMA stuff and disable by default ('numa=on' enables it).
Signed-off-by: Keir Fraser <keir@xensource.com>

Print the time to xend-debug.log, when starting or restarting Xend.

Signed-off-by: Ewan Mellor <ewan@xensource.com>

PV-to-HVM: Add a compatibility define for the end_that_request_last
which changed prototype in 2.6.16.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Include compatibility kzalloc implementation for kernels
before 2.6.14.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-to-HVM: Impletement compatibility version of
schedule_timeout_interruptible for kernels before 2.6.14

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Implement and export a compatibility version of do_exit().
Kernels prior to 2.6.12 did not export do_exit().

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Implement compatibilty version of wait_for_completion_timeout
for kernels before 2.6.11

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>

PV-on-HVM: Include compatability vmalloc_to_pfn for kernels before 2.6.10

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Kernels prior to 2.6.8 did not export strcspn to modules
therefore implement our own and export it.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Define system_state as 1 (SYSTEM_RUNNING) on kernels prior
to 2.6.7. These kernels did not export the system_state variable to
modules.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>

PV-to-HVM: Implement compatibility nonseekable_open as a nop on
kernels before 2.6.9

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>

PV-to-HVM: Add compatibility definitions of various #defines when they
are not present in the kernel we are building against.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Include compatability headers for asm/pgtable-no{pmd,pud}.h
for kernels before 2.6.11.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Add a compatibility linux/mutex.h for kernels before
2.6.16.  This implements the new mutex type in terms of the old
semaphore type.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Add a compatibility linux/io.h for kernels before 2.6.16

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

PV-on-HVM: Add infrastructure for a compatibility shim in order to
support PV-on-HVM drivers for older kernel releases.

The compatability code is included in the xen-platform-pci module
since this module underpins all the others.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] PV-on-HVM: The elevator_init prototype changed in 2.6.10

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] PV-on-HVM: Implement HDIO_GETGEO ioctl on kernels before
2.6.16.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] PV-on-HVM: The probe, remove and shutdown methods were added
to struct bus_type in 2.6.16. For older kernels use the methods on the
struct device_driver.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] PV-on-HVM: Only initialize the owner field of struct
device_driver on kernels from 2.6.10 onward. The field was not
available until then.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] Exclude support for /proc/xen/xsd_kva when building a
non-privileged guest.

This is useful for PV-on-HVM because it removes code which would
otherwise require extra code in the compatability shim.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>

[LINUX] PV-on-HVM: Remove the attr parameter from device attribute
show methods on kernels prior to 2.6.13. The parameter was not present
before that.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] Move xenbus backend probing into a separate source file and
only build it when CONFIG_XEN_BACKEND is set.

This removes unused code from frontend only configurations and also
makes PV-on-HVM drivers on older kernels simpler by removing code
which would otherwise require extra code in the compatability shim.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>

PV-on-HVM: Include a Makefile in each subdirectory to maintain
compatability with kernels from before Kbuild files were introduced.

This follows best practice as described in
Documentation/kbuild/modules.txt

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>

[LINUX] PV-on-HVM: Directly include headers which are used in various
files rather than relying on them being included indirectly.

This is required because these headers are not pulled in indirectly on
older kernels.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>

[XEN] Make memory hypercalls NUMA-aware.

This patch modifies memory ops to use the NUMA-aware page allocator
functions.  We use the target domain's VCPU0 placement to determine
which node's memory to use.  We expect the system administrator to
utilize the exposed NUMA topology information to help craft guest
config files that are NUMA-friendly (use only processors and memory values
that will fit within a given node).

Signed-off-by: Ryan Harper <ryanh@us.ibm.com>

[XEN] NUMA-ify Xen heap and page allocator

This patch adds a per-node bucket to the heap structure in Xen.
During heap initialization the patch determines which bucket to place
the memory. We reserve guard pages between node boundaries in the case
that said boundary isn't already guarded by the MAX_ORDER boundary to
prevent the buddy allocator from merging pages between nodes.

Signed-off-by: Ryan Harper <ryanh@us.ibm.com>

[XEN] Add basic NUMA/SRAT support to Xen from Linux 2.6.16.29.
Signed-off-by: Ryan Harper <ryanh@us.ibm.com>

merge

[HVM] Move ACPI timer to HV, which is the last platform timer in Qemu.

We found Vista polls ACPI timer very frequently (about 15 times
averagely) when handling timer (RTC @ 64HZ) interrupt routine. Though
the exact reason is known, it should be related to system time
adjustment. When it's in Qemu, the overhead is big. After moving,
Vista's idle overhead decreases dramatically from ~10% to 0.9%.=20

Another benefit is that Vista can only pass Performance rating with
this patch. The root cause is that ACPI timer in Qemu isn't
synchronous with other platform timer in HV, which results in Vista
complains "can't measure TSC frequency".

This patch changes vpit.h to vpt.h, for it not only has pit structure
in it, but other platform timer's structure. Another change is moving
ACPI timer and related address from acpi.h to ioreq.h, which can be shared
by HV and ACPI firmware.

Signed-off-by: Xiaowei Yang <xiaowei.yang@intel.com>

[XEND] Be more generous with the default shadow allocation.
Allow enough shadow memory to avoid thrashing the shadow pages,
rather than just enough for safety.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>

[HVM] Avoid buffer overrun in qemu-dm
The array offset in set_bits_in_row here comes from an otherwise un-checked
VNC client request.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>

[BLOCK] blkback: Fix potential grant entry leaks on error

As it stands grant entries for the actual data are mapped in bulk.
If one of the earlier entries fail to be mapped, it can cause all
subsequent entries to be leaked if they were successfully mapped.

This patch changes it so that we will continue to reap grant entries
even when an error is detected.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[BLKTAP] Simplify linked-list insertion/deletion.
This should also avoid (bogus) compiler warnings, as reported
on IA64.
Signed-off-by: Keir Fraser <keir@xensource.com>

[XEND] Open xend-debug.log in append mode.
Signed-off-by: John Levon <john.levon@sun.com>

[TOOLS][XENTOP] Update usage and manpage to include vbd options.

Also a couple of whitespace fixes.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

merge with xen-unstable.hg

[IA64] xenctx shows more registers for ia64

This patch adds more user registers to show them to xenctx for ia64.
Tested domU/domVTi on ia64.

Sample is the below.
# ./xenctx 1 0
 iip:               e000000000000810
 ipsr:              00001012087a6010   b0:                a000000100068a70
 b6:                a00000010014ff60   b7:                e000000000000800
 cr_ifs:            800000000000050a   ar_unat:           0000000000000000
 ar_pfs:            8000000000000209   ar_rsc:            0000000000000008
 ar_rnat:           0000000000000000   ar_bspstore:       a000000100c19030
 ar_fpsr:           0009804c8a70433f   event_callback_ip: a000000100067a20
 pr:                000000000005aa85   loadrs:            0000000000780000
 iva:               a000000100008000   dcr:               0000000000007e04

 r1:  a0000001010369a0
 r2:  0000000000001000   r3:  8000000000000209
 r4:  0000000000000000   r5:  0000000000000000
 r6:  0000000000000000   r7:  0000000000000000
 r8:  a000000100068a70   r9:  0000000000000100
 r10: 0000000000000000   r11: 0000000000050ac5
 sp:  a000000100c1fd80   tp:  a000000100c18000
 r14: 0000000000000001   r15: 0000000000000000
 r16: fffffffffff04c18   r17: a000000100c1fdb0
 r18: a000000100c1fdb1   r19: a000000100c1fe90
 r20: a000000100c1fe10   r21: 0000000000000000
 r22: 0000000000000001   r23: 0000000000000000
 r24: a000000100e5a448   r25: a000000100c18f10
 r26: ffffffffffff0030   r27: 0000000000000000
 r28: 000000000000001d   r29: 0000000000000000
 r30: 0000000000000000   r31: 0000000000000000

 itr: P rid    va               pa            ps      ed pl ar a d ma    key
 [0]  1 000005 a000000100000000 0000004000000 1a  64M 1  2  3  1 1 0 WB  000000
 [1]  1 000007 e000000000000000 0000000000000 18  16M 1  2  3  1 1 0 WB  000000
 [2]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [3]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [4]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [5]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [6]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [7]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000

 dtr: P rid    va               pa            ps      ed pl ar a d ma    key
 [0]  1 000005 a000000100000000 0000004000000 1a  64M 1  2  3  1 1 0 WB  000000
 [1]  1 000007 ffffffffffff0000 0000000010000 10  64K 1  2  3  1 1 0 WB  000000
 [2]  1 000007 e000000000000000 0000000000000 18  16M 1  2  3  1 1 0 WB  000000
 [3]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [4]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [5]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [6]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000
 [7]  0 000000 0000000000000000 0000000000000 00      0  0  0  0 0 0 WB  000000

Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>

[IA64] Support multiple page sizes in VHPT

Enable VHPT support for multiple page sizes.

Signed-off-by: Anthony Xu <anthony.xu@intel.com>

Add x86_64 to the supported architectures since otherwise the tests
don't run.

Signed-off-by: Stefan Berger <stefanb@us.ibm.com>

[IA64] Move console_start_sync() in INIT path

This patch moves console_start_sync() before first message in
ia64_init_handler(), and it cleans up ia64_init_handler().

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>

Fix Memory assumptions in the create tests.

Use the architecture specified idea of minimum memory.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

[qemu patches] Update patches upto changeset 11870:29b02d929b7e.

Signed-off-by: Christian Limpach <Christian.Limpach@xensource.com>

This puts two small fixes in the xm-test suite:
- it turns off verbose mode of the ACM module
- copies the test policy to the policies directory only if the directory
has been created

Signed-off-by: Stefan Berger <stefanb@us.ibm.com>

Added multiarch support to xm-test.  Bumped version number to xm-test 1.0.
New README for public ramdisks.

Signed-off-by: Ewan Mellor <ewan@xensource.com>

Update buildroot configs to match snapshot downloaded for public ramdisks.

Signed-off-by: Ewan Mellor <ewan@xensource.com>

Ignore more build-time files for xm-test.

Signed-off-by: Ewan Mellor <ewan@xensource.com>

Default to appending to "extra" in XenConfig.

PowerPC needs console information from the command line.  Resetting the whole
command line causes false failures.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

Ignore generated .test files.

Signed-off-by: Jimi Xenidis <jimix@watson.ibm.com>
Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

Add ability to inspect messages from domain for arbitrary strings.

Used on PowerPC to FAIL(), if the guest domain hits a BUG() and enters XMON.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

Add configuration data for powerpc.

Allows for building of initrd, and running of tests.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

Refactor code to encapsulate architecture decisions in one place.

Also includes some whitespace fixes.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

Update .hgignore to remove artifacts of ramdisk build.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

Rename buildroot -> buildroot-i386.

Also update Makefile.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>

[XEN] Remove definition of printf. All users are switched to printk.
Signed-off-by: Keir Fraser <keir@xensource.com>

[BLKTAP] Do not allocate a descriptor for blktap control device
(minor 0).

Signed-off-by: Steven Rostedt <srostedt@redhat.com>

[BLKTAP] Only allocate tapfd descriptors when they are requested.

Currently all are allocated at bootup, even when they will never be used.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>

[BALLOON] Add a sysfs interface.
Information is located under /sys/devices/system/memory.

Original patch from Satoshi UCHIDA <s-uchida@ap.jp.nec.com>
Signed-off-by: Keir Fraser <keir@xensource.com>

[XEN] Fill in runstate_info when virtual address is registered,
even when it is registered from the 'wrong' VCPU.
Signed-off-by: Keir Fraser <keir@xensource.com>

[XM] Print qualified path to config file on domain creation.
From: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Keir Fraser <keir@xensource.com>

Allow domain config file to specify credit-scheduler parameters.
Signed-off-by: Masami Watanabe <masami.watanabe@jp.fujitsu.com>

[XM] Allow empty resource strings (e.g., when creating an empty CDROM
device). Also clean up error/acces-denied path.
Signed-off-by: Keir Fraser <keir@xensource.com>

[HVM] No need to call hvm_do_resume() on every vm entry.
Original patch from Xin Li <xin.b.li@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>

[NET] back: Fix packet queuing so that packets are drained if the
interface is blocked for longer than 500ms. This avoids deadlock
situations where interfaces cannot be destroyed because some other
dormant interface is holding resources.

Signed-off-by: Christopher Clark <christopher.clark@cl.cam.ac.uk>
Signed-off-by: Keir Fraser <keir@xensource.com>

[BLOCK] blkback: Fix first_sect check.

The blkback code checks first_sect by computing last_sect - first_sect
+ 1. If this is not positive then the request is
rejected. Unfortunately, the calculation is stored in an unsigned
integer before the check is made so negative values will not be
rejected.

This patch fixes this by explicitly comparing first_sect with
last_sect.

The blktap code is similar but it stores the difference in a signed
integer so it's okay.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[BLOCK] blkback: Copy shared data before verification

As it is blkback verifies the metadata from the frontend in place.
This means we run the risk of the frontend changing the data after
we've verified it.  This patch copies the data onto the stack before
verifying and using it to ensure we see a consistent snapshot.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[ACM] Don't use uninitialised struct value if hypercall fails.
From: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Keir Fraser <keir@xensource.com>

[SOLARIS] Don't mlock() hypercall buffers.

On solaris mlock requires a page aligned address and mlock doesn't
ensure the pages won't minor page fault.

Signed-off-by: Mark Johnson <mark.johnson@sun.com>

merge with xen-unstable.hg

[IA64] Fix Windows Timer stop issue.

When doing HCT testing on Guest Windows, Windows Timer might stop,
if there is no mouse activity.

When implementing vhpi acceleration, I didn't think of below situation.
windows uses "epc" instruction to implement system call like linux,
In a very small code sequence including "epc" instruction, if there is
an external interrupt, Windows will defer handling this external interrupt
by rfi with ipsr.i=0.

Signed-off-by, Anthony Xu <anthony.xu@intel.com>

[IA64] remove unused vmx/mm.c file

Signed-off-by: Tristan Gingold <tristan.gingold@bull.net>

[IA64] BUG() bad path during domain destruction

BUG out on badness due to SMP problems accessing the p2m table.  This
is temporary until we have a real fix.  To avoid this, do not use the
RX copy option (xennet.rx_copy).

Based on patch by: Tsunehisa Doi

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

[IA64] sync_console in ia64_init_handler

This patch fix the following issue.

1. boot xen
2. push INIT bottun
3. Nothing is printed to serial console.

I add console_start_sync() into ia64_init_handler().
Then this issue is fixed.

Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>

Free correct number of maptack pages during domain destruction.
Signed-off-by: Christopher Clark <christopher.clark@cl.cam.ac.uk>

Use $CC when building pygrub.
Signed-off-by: John Levon <john.levon@sun.com>

[HVM] Replace hvm_prod_vcpu() with calls to generic (and correct) vcpu_kick().
Signed-off-by: Keir Fraser <keir@xensource.com>

[HVM] Don't compare CR3 pfn to domain->max_pages
max_pages is the largest number of pages the domain can have
allocated, not the gpfn of the highest one.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>

[XEN] Stricter TLB-flush discipline when unshadowing pagetables
It's OK for the guest to see old entries in the TLB, but not for the
shadow fault handler to see them in its linear mappings.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>

[XEN] Show shadow status of vcpus in domain info dump
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>

Omit seg fixup warning if it's just from statically linked init.

Signed-off-by: Steven Hand <steven@xensource.com>

Workaround for bigsmp APIC mode SMP linux guests - we cannot correctly deal
with timer interrupts being delivered to VCPUs other than 0, so spit a
warning and then subsequently ignore the target. This fixes the recent
issues Intel reported when booting SMP FC5 and FC6 kernels.

Needs a better fix at some point.

Signed-off-by: Steven Hand <steven@xensource.com>

[XEND] Remove lazy import of signal module.
Signed-off-by: Jeremy Katz <katzj@redhat.com>

Allow default target architecture of PV-on-HVM drivers to be overridden.
Signed-off-by: Travis Betak <travis.betak@amd.com>

[ACM] Documentation cleanup.

This patch eliminates redundant security tools information that was
integrated into the Xen user guide and the xm man page.

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

[NET] gso: Fix rcv mss estimate

I noticed that with default TCP window sizes TSO + Xen would slow to a
crawl on certain machines.  It turned out that there is a bug in the
TCP stack when it comes to receiving LRO (counter part of TSO for rx)
packets.

The following fix has been applied upstream.

        [TCP]: Fix rcv mss estimate for LRO

        By passing a Linux-generated TSO packet straight back into
        Linux, Xen
        becomes our first LRO user :) Unfortunately, there is at least
        one spot
        in our stack that needs to be changed to cope with this.

        The receive MSS estimate is computed from the raw packet
        size.  This is
        broken if the packet is GSO/LRO.  Fortunately the real MSS can
        be found
        in gso_size so we simply need to use that if it is non-zero.

        Real LRO NICs should of course set the gso_size field in
        future.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[HVM] Fix oprofile for HVM guests.

Signed-off-by: Tom Woller <thomas.woller@amd.com>
Signed-off-by: Mats Petersson <mats.petersson@amd.com>
Signed-off-by: Keir Fraser <keir@xensource.com>

[HVM][SVM] Avoid double-execution of NMI handler.

Remove the do_nmi() call from the vmexit logic, as the NMI is held
pending and will be handled in the context of the host when the STGI
instruction is executed.

Signed-off-by: Tom Woller <thomas.woller@amd.com>

[HVM][SVM] Fix 2/2 to nterrupt delivery logic.

Add flag to indicate that an exception event needs injecting, and to
delay the ext interrupt injection.
Remove unnecessary check of RFLAGS.IF for ExtInt injection.

Signed-off-by: Travis Betak <travis.betak@amd.com>
Signed-off-by: Wei Huang <wei.huang2@amd.com>
Signed-off-by: Tom Woller <thomas.woller@amd.com>

[HVM][SVM] Fix 1/2 to interrupt delivery logic.

This patch uses the VINTR intercept to signal the hypervisor when a
guest can take an interrupt.  When guest's interrupts are masked by
EFLAGS.IF or the guests are in an interrupt shadow, we create a 'fake'
virtual interrupt to inject while also enabling the VINTR intercept.
When the guest _can_ take interrupts, the hypervisor will #VMEXIT
on VINTR. The VINTR exit handler then clears the VINTR intercept bit
and clears the V_IRQ bit so that svm_intr_assist() can inject a
legitimate interrupt.

Signed-off-by: Travis Betak <travis.betak@amd.com>
Signed-off-by: Wei Huang <wei.huang2@amd.com>
Signed-off-by: Tom Woller <thomas.woller@amd.com>

[HVM] Implement password authentication of VNC connections.

The specification is as mentioned at
http://lists.xensource.com/archives/html/xen-devel/2006-09/msg00666.html
(However, password came to describe plain text)

The difference is follows.
- protocol_authtype() without the necessity was deleted.
- The check on the protocol version was added.
- And, some small modification.

Signed-off-by: Masami Watanabe <masami.watanabe@jp.fujitsu.com>

Use the name of the title of the system to boot into (instead of the
kernel version) to determine where to make the entry into the grub
configuration file.

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

[HVM] Pass all non-RAM memory accesses to handle_mmio().

It is up to handle_mmio() to correctly handle accesses outside
valid device mmio regions.

Signed-off-by: Keir Fraser <keir@xensource.com>

[BLKTAP]: Fix potential grant entry leaks on error

As it stands grant entries for the actual data are mapped in bulk.
If one of the earlier entries fail to be mapped, it can cause all
subsequent entries to be leaked if they were successfully mapped.

This patch changes it so that we will continue to reap grant entries
even when an error is detected.

There is still a potential leak if create_lookup_pte_addr fails
in fast_flush_area.  This leak is not affected by this patch.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[BLKTAP]: Kill bogus fast_flush_area

The function dispatch_rw_block_io may call fast_flush_area even before
a pending request has been initialised.  This patch deletes that call.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[BLKTAP]: Kill duplicate fast_flush_area call

The dispatch_rw_block_io may call fast_flush_area twice if
create_lookup_pte_addr fails (there is a flush call at fail_flush
already).  The second call simply causes warnings to be printed
on the console.  This patch removes the duplicate call.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

[VMXASSIST] Some reports that macros do not get picked up by CPP
if they are prefixed with $. Fix by placing macro uses in ().
Signed-off-by: Keir Fraser <keir@xensource.com>