Keir Fraser [Sat, 24 Feb 2007 12:19:22 +0000 (12:19 +0000)]
xen memory allocator: Dynamically allocate per-numa-node metadata.
Signed-off-by: Keir Fraser <keir@xensource.com>
Jake Wires [Sat, 24 Feb 2007 01:26:07 +0000 (17:26 -0800)]
[TAPDISK] honor read-only attributes when creating tap-based VBDs
Signed-off-by: Jake Wires <jwires@xensource.com>
Jake Wires [Fri, 23 Feb 2007 17:40:32 +0000 (09:40 -0800)]
[TAPDISK] handle error case when opening VDIs
kfraser@localhost.localdomain [Fri, 23 Feb 2007 17:02:58 +0000 (17:02 +0000)]
xen memory alloctor: remove bit width restrictions
Hide the (default or user specified) DMA width from anything outside
the heap allocator. I/O-capable guests can now request any width for
the memory they want exchanged/added.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 17:01:38 +0000 (17:01 +0000)]
xen memory allocator: per-bit-width heap zones
Replace the 3-zone scheme of the heap allocator with one with one
where zones are distinguished by their bit widths.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Ian Campbell [Fri, 23 Feb 2007 16:57:34 +0000 (16:57 +0000)]
merge
Ian Campbell [Fri, 23 Feb 2007 16:56:45 +0000 (16:56 +0000)]
[LINUX] Purge include <linux/config.h>. It has been obsolete for some time now.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Ian Campbell [Fri, 23 Feb 2007 16:56:42 +0000 (16:56 +0000)]
[PV-on-HVM] Fix gnttab_map to not run off the bottom of the gnttab space.
Cleanup a few warnings: reached end of void function and unused variables.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 16:55:56 +0000 (16:55 +0000)]
xen memory allocator: hide generic allocator routines
This patch doesn't introduce functional changes, but simply moves code
around to make the unused (outside of the page allocator) heap alloc
functions taking an explicit zone parameter static without having to
forward-prototype them in their source file.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 16:36:55 +0000 (16:36 +0000)]
xentrace: Add a relative TSC field (kept track per CPU) to xentrace_format.
Additionally a default line is added (key 0x00000000) which is used
when no specific entry in the format file is found.
Signed-off-by: Thomas Friebel <thomas.friebel@amd.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 16:24:07 +0000 (16:24 +0000)]
linux, tpm: Fix argument to grant table operation
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 16:17:00 +0000 (16:17 +0000)]
hvm: Fix save/restore to handle missing pages.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 16:16:36 +0000 (16:16 +0000)]
hvm: unset_mm_mapping() for vga acceleration region should recalculate
maxmem setting before freeing memory!
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Fri, 23 Feb 2007 14:44:07 +0000 (14:44 +0000)]
[HVM] Only treat gfns in known MMIO regions as MMIO
instead of assuming that all non-RAM addresses are MMIO
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 14:36:33 +0000 (14:36 +0000)]
xen: Fix dumb boot allocator breakage that prevents loop termination
in some cases. :-)
Signed-off-by: Keir Fraser <keir@xensource.com>
Emmanuel Ackaouy [Fri, 23 Feb 2007 12:58:20 +0000 (12:58 +0000)]
[XEN] [TOOLS] 32on64 compat save/restore: unadjust L3E flags at save time
Get rid of USER|RW|ACCESSED L3 PTE flags which may be present
in a PAE guest when running on a 64bit hypervisor at save time.
This prevents the restore operation from getting confused by
these flags.
Reset disallowed L3E flags to match PAE case.
Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 11:32:25 +0000 (11:32 +0000)]
hvm: Clean up initialisation of hvm_funcs.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 10:47:55 +0000 (10:47 +0000)]
Implement VBD and VIF handling on Solaris for libxenstat.
Also, fix up a confusion with ERR that was breaking xentop.
Signed-off-by: John Levon <john.levon@sun.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 10:43:00 +0000 (10:43 +0000)]
linux: Fix current_vcpu_info for UP case
Under !CONFIG_SMP, smp.h is not always included and hence
smp_processor_id() not always visible.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
kfraser@localhost.localdomain [Fri, 23 Feb 2007 10:38:33 +0000 (10:38 +0000)]
Merge with xen-ia64-unstable.hg.
kfraser@localhost.localdomain [Fri, 23 Feb 2007 10:35:16 +0000 (10:35 +0000)]
hvm: Rename injection_pending() to event_injection_faulted().
Fix the VMX and SVM handlers to reflect the new semantics (which is
what is actually required by the one caller, in shadow fault path).
Signed-off-by: Keir Fraser <keir@xensource.com>
Jake Wires [Fri, 23 Feb 2007 05:32:17 +0000 (21:32 -0800)]
[TAPDISK] minor changes to tapdisk and plugins
* open all parent images read-only
* expose parent image names to tapdisk (needed for locking api)
Signed-off-by: Jake Wires <jwires@xensource.com>
kfraser@localhost.localdomain [Thu, 22 Feb 2007 22:27:40 +0000 (22:27 +0000)]
x86/64: Fix a user of alloc_boot_pages() who should use
alloc_boot_low_pages() instead.
Signed-off-by: Keir Fraser <keir@xensource.com>
awilliam@xenbuild2.aw [Thu, 22 Feb 2007 17:15:29 +0000 (10:15 -0700)]
merge with xen-unstable.hg
awilliam@xenbuild2.aw [Thu, 22 Feb 2007 16:42:13 +0000 (09:42 -0700)]
[IA64] First step to support big-endian PV domUs
Signed-off-by: Dietmar Hahn <dietmar.hahn@fujitsu-siemens.com>
kfraser@localhost.localdomain [Thu, 22 Feb 2007 15:26:21 +0000 (15:26 +0000)]
alloc_boot_pages() allocates downwards from high memory.
This conserves low memory.
Provide new function alloc_boot_low_pages() for those callers who
actually require lowmem pages (e.g., below 4GB).
Based on a patch by Chris Lalancette <clalance@redhat.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 22 Feb 2007 15:22:16 +0000 (15:22 +0000)]
x86: Detect shadow-emulate write to stack from inside the write
handler, where we can quite conveniently check whether the access
segment is SS. This will only give a false positive for non-stack
writes that override the destination segment to SS, which it is
probably safe to assume will never happen when the destination is a
legitimate page table!
For now, instead of bailing and unshadowing just increment a perfctr
and we'll see if that increases fast under any reasonable workload.
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Thu, 22 Feb 2007 14:58:26 +0000 (14:58 +0000)]
Fix perfc=y build
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Thu, 22 Feb 2007 13:03:49 +0000 (13:03 +0000)]
hvm: Turn stack-pointer comparison on shadow-emulate path into a
heuristic which we can allow to fail.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Thu, 22 Feb 2007 12:49:44 +0000 (12:49 +0000)]
hvm: Do not take shadow-emulation path if we are trying to inject an
event into the HVM guest.
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Thu, 22 Feb 2007 12:21:29 +0000 (12:21 +0000)]
[XEN] don't emulate stack operations on shadowed pages
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Jake Wires [Thu, 22 Feb 2007 02:38:11 +0000 (18:38 -0800)]
[BLKTAP] fix sysfs cleanup on error case
Signed-off-by: Jake Wires <jwires@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 20:30:37 +0000 (20:30 +0000)]
x86: Only unshadow on failed emulation, not when an exception is raised.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 20:03:56 +0000 (20:03 +0000)]
linux: Use hard tabs and Linux coding style.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 19:58:58 +0000 (19:58 +0000)]
x86: Extend emulator return codes.
Signed-off-by: Keir Fraser <keir@xensource.com>
Jake Wires [Wed, 21 Feb 2007 19:42:04 +0000 (11:42 -0800)]
[BLKBACK/BLKTAP] add sysfs throughput profiling to blk{back/tap}
kfraser@localhost.localdomain [Wed, 21 Feb 2007 19:12:16 +0000 (19:12 +0000)]
hvm: local_events_need_delivery() must take into account VLAPIC/PIC status.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 18:54:48 +0000 (18:54 +0000)]
libfs: Fix build with gcc4.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 16:33:34 +0000 (16:33 +0000)]
Add find_domain_rcu_lock() and domain_rcu_unlock().
Signed-off-by: Jose Renato Santos <jsantos@hpl.hp.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 16:13:49 +0000 (16:13 +0000)]
Use RCU for domain_list and domain_hash.
Signed-off-by: Jose Renato Santos <jsantos@hpl.hp.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Wed, 21 Feb 2007 14:44:09 +0000 (14:44 +0000)]
Merge
kfraser@localhost.localdomain [Wed, 21 Feb 2007 14:40:37 +0000 (14:40 +0000)]
x86: Better BUG() and ASSERT() logging.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Wed, 21 Feb 2007 10:13:40 +0000 (10:13 +0000)]
x86: Allow exceptions to be handled while interrupts are
disabled. Handlers must take special care if necessary.
Fixes the debug 'd' key.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ewan Mellor [Wed, 21 Feb 2007 00:04:59 +0000 (00:04 +0000)]
Added C bindings for host.supported_bootloaders.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Ewan Mellor [Wed, 21 Feb 2007 00:04:06 +0000 (00:04 +0000)]
Fix allocation of the error_description fields.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 23:59:13 +0000 (23:59 +0000)]
Put the set size in the structure on set-allocation.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 22:54:30 +0000 (22:54 +0000)]
Added host.supported_bootloaders field.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 22:46:21 +0000 (22:46 +0000)]
Remove host.create and host.destroy -- these never made any sense.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
awilliam@xenbuild2.aw [Tue, 20 Feb 2007 22:43:57 +0000 (15:43 -0700)]
merge with xen-unstable.hg
Ewan Mellor [Tue, 20 Feb 2007 22:40:55 +0000 (22:40 +0000)]
Added modelling and C bindings for VIF/VBD QoS parameter setting. No Xend
implementation yet.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
awilliam@xenbuild2.aw [Tue, 20 Feb 2007 22:12:11 +0000 (15:12 -0700)]
[IA64] Update xencommize_mini_grant_table_op for GNTTABOP_query_size
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Ewan Mellor [Tue, 20 Feb 2007 21:55:11 +0000 (21:55 +0000)]
Split the VIF I/O statistics fields out into a VIF_metrics class. Implement
this in Xend, and fix the VBD_metrics implementation. Implement C bindings
also.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 21:28:19 +0000 (21:28 +0000)]
Split the VBD I/O statistics out into a VBD_metrics class. Document the
VBD.type field, and add it to the C bindings (this was already in Xend).
Signed-off-by: Ewan Mellor <ewan@xensource.com>
awilliam@xenbuild2.aw [Tue, 20 Feb 2007 19:58:22 +0000 (12:58 -0700)]
merge with xen-unstable.hg
Ewan Mellor [Tue, 20 Feb 2007 19:38:24 +0000 (19:38 +0000)]
Added Console.other_config to the docs (Xend already has this).
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 19:23:28 +0000 (19:23 +0000)]
Rename Console.uri to location. Xend already uses the latter.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 17:05:50 +0000 (17:05 +0000)]
x86 emulation: fix bswap
REX.R needs to be decoded. Since it is simple to do, faster, and
smaller, let hardware do 32- and 64-bit BSWAP. 16-bite BSWAP is
undefined: emulate with observed behaviour (write zero).
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 16:57:50 +0000 (16:57 +0000)]
x86: Fix emulation of REP prefix.
Firstly, it should be ignored when used with any opcode for which it
is undefined. Secondly, the count register (rCX) width depends on
address size.
Signed-off-by: Keir Fraser <keir@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 16:30:24 +0000 (16:30 +0000)]
Remove VM.VCPUs_number and VM.VCPUs_utilisation as these have moved onto the
VM_metrics class. Remove VM_metrics.VM -- the backreference is unnecessary
and awkward to deal with.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
Tim Deegan [Tue, 20 Feb 2007 16:19:40 +0000 (16:19 +0000)]
Merge
Tim Deegan [Tue, 20 Feb 2007 16:18:56 +0000 (16:18 +0000)]
[HVM] Another unchecked caller of hvm_copy_to_guest_virt()
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 16:12:02 +0000 (16:12 +0000)]
acm: More fixes after grant-table changes. These are all bugs
introduced during merge of Isaku's previous patch. :-)
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
Tim Deegan [Tue, 20 Feb 2007 15:37:28 +0000 (15:37 +0000)]
Merge
Tim Deegan [Tue, 20 Feb 2007 15:37:03 +0000 (15:37 +0000)]
[XEN] Get rid of gva_to_gpa translation
It didn't have any sensible error checking. Make all callers
use gva_to_gfn translation and check the result. MMIO and PIO
callers inject pagefaults to the guest iof the non-IO address is
not mapped.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Ewan Mellor [Tue, 20 Feb 2007 15:24:01 +0000 (15:24 +0000)]
Replace the X_HANDLE_INVALID error codes with a single HANDLE_INVALID that
names the class as a parameter.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 23:54:42 +0000 (23:54 +0000)]
hvm: Fix getvcpucontext to return the correct CR3 value.
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Tue, 20 Feb 2007 20:49:44 +0000 (20:49 +0000)]
[HVM] Fix MMIO AND emulation
which was falling though into AND...
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Tim Deegan [Tue, 20 Feb 2007 20:02:49 +0000 (20:02 +0000)]
[HVM] Fix MMIO LODS emulation
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
Vincent Hanquez [Tue, 20 Feb 2007 14:12:12 +0000 (14:12 +0000)]
add /sbin to the $PATH and use ldconfig -p instead of ldconfig -v
Tim Deegan [Tue, 20 Feb 2007 13:57:26 +0000 (13:57 +0000)]
[XEN] Fix skip-verify shadow heuristic
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 12:27:03 +0000 (12:27 +0000)]
linux: Fix swiotlb bug-out due to uninitialised end_pfn/max_mapnr.
Signed-off-by: Keir Fraser <keir@xensource.com>
Vincent Hanquez [Tue, 20 Feb 2007 12:22:40 +0000 (12:22 +0000)]
add /sbin to the $PATH for debian's ldconfig.
use ldconfig -p instead of -v that doesn't require root privilege to work.
kfraser@localhost.localdomain [Tue, 20 Feb 2007 12:18:29 +0000 (12:18 +0000)]
linux/x86: Advertise which page table entries contain MFNs and hence
need to be (un)canonicalized during save/restore.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 11:57:06 +0000 (11:57 +0000)]
xen/x86: Better BUG back traces.
Signed-off-by: Keir Fraser <keir@xensource.com>
Tim Deegan [Tue, 20 Feb 2007 11:51:40 +0000 (11:51 +0000)]
[XEN] Fix potential race in unshadow code
And add some diagnostic printout in case it happens again
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
kfraser@localhost.localdomain [Tue, 20 Feb 2007 09:41:32 +0000 (09:41 +0000)]
linux/x86: Clean up page table handling headers
- remove dead code
- fix line breaking and space vs. tab usage
- remove redundant parentheses
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Tim Deegan [Tue, 20 Feb 2007 09:34:22 +0000 (09:34 +0000)]
Merge
kfraser@localhost.localdomain [Tue, 20 Feb 2007 09:33:14 +0000 (09:33 +0000)]
linux/x86: Adjust and fix page table handling.
Ensure that all and only those page table entries that have their
present bit set undergo p2m/m2p translation in all relevant places.
This should fix migration issues with _PAGE_PROTNONE pages which
previously could retain MFNs in PTEs while having the present bit
clear (and thus were not getting (un)canonicalized during save/
restore).
Many thanks to Keir Fraser for his analysis, suggestions, and
corrections.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
john.levon@sun.com [Tue, 20 Feb 2007 04:44:42 +0000 (20:44 -0800)]
Xenconsoled should ignore spurious watch event. Otherwise, it can rebind to the same evtchn of a dying domU during suspending and cause below error message:
(XEN) event_channel.c:177:d0 EVTCHNOP failure: domain 0, error -22, line 177
john.levon@sun.com [Tue, 20 Feb 2007 04:44:42 +0000 (20:44 -0800)]
Filesystem implementations may need optional arguments in terms of
what to mount. Add an options string to the libfsimage API.
Signed-off-by: John Levon <john.levon@sun.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 22:50:07 +0000 (22:50 +0000)]
linux: Fix 32-bit build after changeset 13993:
168030c8e0a93d8401cf5e783e10ecdf0938abda
Signed-off-by: Keir Fraser <keir@xensource.com>
Emmanuel Ackaouy [Mon, 19 Feb 2007 19:58:07 +0000 (19:58 +0000)]
[XEN] 32on64 shadowing / live migration support for PV PAE compat guests
PAE compat guests on 64bit hypervisors are shadowed
using 4-on-4 with special handling for the top level
L4 page and the L2E M2P mappings.
Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 16:16:53 +0000 (16:16 +0000)]
Quieten noisy fied-up-GPF-in-Xen logging.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 16:11:56 +0000 (16:11 +0000)]
linux: Introduce {current_,}vcpu_info.
Also consolidate time-xen.c in giving get_time_values_from_xen() a
'cpu' parameter to match other functions (this consolidation could of
course also be done the other way around, but I think this way the
resulting code can be more efficient).
Signed-off-by: Jan Beulich <jbeulich@novell.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 16:06:03 +0000 (16:06 +0000)]
ia64: Fixes after changes to grant table code.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 16:05:34 +0000 (16:05 +0000)]
acm: Further fixes after grant-table changes.
Based on a patch from Isaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 15:52:51 +0000 (15:52 +0000)]
Fix ACM build after recent grant-table code changes.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 15:51:46 +0000 (15:51 +0000)]
hvm, vtpm: Fix wait for an outstanding response.
Post the interrupt upon reception of the response and adapt the
state machine.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 15:50:29 +0000 (15:50 +0000)]
minios: Fix xenbus-id allocation loop.
Signed-off-by: Dietmar Hahn <dietmar.hahn@fujitsu-siemens.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 15:46:01 +0000 (15:46 +0000)]
hvm: Remove noisy warnings associated with MOVS emulation.
Signed-off-by: Keir Fraser <keir@xensource.com>
kfraser@localhost.localdomain [Mon, 19 Feb 2007 15:43:27 +0000 (15:43 +0000)]
hvm: Emulate opcode 0x03 (ADD mem,reg).
Signed-off-by: Keir Fraser <keir@xensource.com>
Ian Campbell [Mon, 19 Feb 2007 12:21:41 +0000 (12:21 +0000)]
[LIBELF] Prefer PT_NOTE segments to SHT_NOTE sections for ELF notes.
It's always an error to try to use sections on an executable; the
segments in the phdr are definitive.
Unfortunately we cannot drop SHT_NOTE support completely due to a
binutils bug which causes kernels to have the offset field of the
PT_NOTE phdr set to zero:
http://sourceware.org/bugzilla/show_bug.cgi?id=594
This bug is present in binutils 2.17 although some distros have
backported the fix.
Therefore we simply prefer a PT_NOTE segment if we find one otherwise
we still use the SHT_NOTE section (and then the old __xen_guest
section).
Based on a patch from Jeremy Fitzhardinge.
Also added XEN_ELFNOTE_HV_START_LOW to readnotes.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
awilliam@xenbuild2.aw [Sun, 18 Feb 2007 23:13:13 +0000 (16:13 -0700)]
[IA64] Make b2-b5 fixed registers
Signed-off-by: Tristan Gingold <tgingold@free.fr>
awilliam@xenbuild2.aw [Sun, 18 Feb 2007 23:08:11 +0000 (16:08 -0700)]
[IA64] Cleanup in vmx.h
Cleanup: remove unused code.
Signed-off-by: Tristan Gingold <tgingold@free.fr>
awilliam@xenbuild2.aw [Sun, 18 Feb 2007 23:06:24 +0000 (16:06 -0700)]
[IA64] Cleanup in vmx
Cleanup: static added, unused code #if'ed, typos.
Signed-off-by: Tristan Gingold <tgingold@free.fr>
awilliam@xenbuild2.aw [Sun, 18 Feb 2007 23:00:52 +0000 (16:00 -0700)]
[IA64] Paravirtualize mmap handlers of /proc/bus/pci
Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
awilliam@xenbuild2.aw [Sun, 18 Feb 2007 22:57:38 +0000 (15:57 -0700)]
[IA64] Import arch/ia64/pci/pci.c into sparse tree
Necessary for paravirtualizing mmap handler for /proc/bus/pci
Signed-off-by: Akio Takebe <takebe_akio@jp.fujitsu.com>
awilliam@xenbuild2.aw [Sun, 18 Feb 2007 22:53:12 +0000 (15:53 -0700)]
[IA64] Quiet VTi domain bootup
Xen shouldn't be printing anything to the console at a visible
log level for normal startup.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Keir Fraser [Sun, 18 Feb 2007 15:29:40 +0000 (15:29 +0000)]
hvm: Support PUSH from mmio area (opcode 0xFF/6).
Also fix a bogus assertion in vlapic device model.
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser [Sat, 17 Feb 2007 13:26:33 +0000 (13:26 +0000)]
linux: vmalloc_sync_all() is not a no-op when running in PAE mode on Xen.
Also vmalloc_sync_all() is only defined for x86 architectures so make
use of it in driver code conditional on CONFIG_X86.
Signed-off-by: Keir Fraser <keir@xensource.com>