xen.git
20 years agoTest for creating a domain with too little memory.
dan@guaranine.beaverton.ibm.com [Wed, 16 Nov 2005 19:33:12 +0000 (20:33 +0100)]
Test for creating a domain with too little memory.

20 years agoMake 01_reboot test more robust.
dan@guaranine.beaverton.ibm.com [Wed, 16 Nov 2005 19:32:36 +0000 (20:32 +0100)]
Make 01_reboot test more robust.
We need to attach a console before rebooting to make sure the domain did
actually boot in the first place.  Also, use a better method for pulling out
the uptime string.

20 years agoMerged.
emellor@leeni.uk.xensource.com [Wed, 16 Nov 2005 12:38:16 +0000 (13:38 +0100)]
Merged.

20 years agoDom0 might not be called Domain-0, so ask for the name in shutdown.py, instead of...
jmb@artemis.home.pimb.org [Wed, 16 Nov 2005 12:34:52 +0000 (13:34 +0100)]
Dom0 might not be called Domain-0, so ask for the name in shutdown.py, instead of hardcoding it

Signed-off-by: Jody Belka <knew (at) pimb (dot) org>
20 years agoRename XENVER_parameters to slightly more informative
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 11:44:12 +0000 (12:44 +0100)]
Rename XENVER_parameters to slightly more informative
XENVER_platform_parameters.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Wed, 16 Nov 2005 11:37:29 +0000 (12:37 +0100)]
Merged.

20 years agoReview checks for Error string when checking that bad arguments are handled
emellor@leeni.uk.xensource.com [Wed, 16 Nov 2005 11:36:47 +0000 (12:36 +0100)]
Review checks for Error string when checking that bad arguments are handled
correctly.  Many checks were looking using output.find("Error") > 1, which is
incorrect, as we do not guarantee that the word "Error" will be first in the
output (never mind the fact that strings are indexed from 0).

All these checks have been changed to compare against -1.  In particular, this
should fix the failure of create_noparm_neg and create_badparm_neg on a machine
with the /etc/xen/xmdefconfig file in place, as xm create issues the
Using config file "/etc/xen/xmdefconfig" diagnostic before reporting the error.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoFix up start_info_t magic string to match the format of
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 11:23:17 +0000 (12:23 +0100)]
Fix up start_info_t magic string to match the format of
XENVER_capabilities strings.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoThis fixes the following scenerio:
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:41:14 +0000 (11:41 +0100)]
This fixes the following scenerio:
1) gmfn X is L2 table, and a shadow for it is set.
2) A write to page X, and the same page would be L1 table
also.(Something like the linear page table) But the shadow for L1 table
is not set.

The code path will be shadow_fault -> l1pte_write_fault. Because X is L2
table, so shadow_mark_va_out_of_sync is called , this function will
allocate a OO entry for this table, and set the writable_pl1e to -1.
Also it alloc a L1 shadow for X through shadow_map_l1_into_current_l2.
When allocating, free_out_of_sync_state will be called to free all OO
entry. However, since the OO entry allocated on
shadow_mark_va_out_of_sync has the writable_pl1e to -1, this OO entry
will never be released.

Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
20 years agoChange ISA_IO_BASE to fix_to_virt(FIX_ISAMAP_BEGIN). It used to be
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:39:38 +0000 (11:39 +0100)]
Change ISA_IO_BASE to fix_to_virt(FIX_ISAMAP_BEGIN).  It used to be
PAGE_OFFSET, which was unaltered from vanilla Linux 2.6.12.6 kernels
and caused isa_memcpy_fromio() to not read physical memory from
0xf0000 - 0xfffff correctly.

http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=393

Signed-off-by: Andrew D. Ball <aball@us.ibm.com>
20 years agoRemove unnecessary cast of kmalloc() result.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:35:26 +0000 (11:35 +0100)]
Remove unnecessary cast of kmalloc() result.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agoAdd definitions of some swiotlb symbols to pci-dma.c.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:34:54 +0000 (11:34 +0100)]
Add definitions of some swiotlb symbols to pci-dma.c.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years ago/dev/mem write function should use copy_from_user, not copy_to_user.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:34:19 +0000 (11:34 +0100)]
/dev/mem write function should use copy_from_user, not copy_to_user.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agomerge.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:33:13 +0000 (11:33 +0100)]
merge.

20 years agoCorrectly export xencons functions to avoid ugly 'extern'
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:32:19 +0000 (11:32 +0100)]
Correctly export xencons functions to avoid ugly 'extern'
declarations.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Wed, 16 Nov 2005 10:32:14 +0000 (11:32 +0100)]
Merged.

20 years agoThe following patch removes every (hopefully) reference to network-limit.
emellor@leeni.uk.xensource.com [Wed, 16 Nov 2005 10:31:57 +0000 (11:31 +0100)]
The following patch removes every (hopefully) reference to network-limit.
Upon further investigation it sets Device parameters that don't exist any
longer, thus causing exceptions.  As it looks like this is completely
obsolete, and no one is actively working on it, we should just rip it out to
end the confusion.

Signed-off-by: Sean Dague <sean@dague.net>
20 years agoFix function that shouldn't be globally defined.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:30:32 +0000 (11:30 +0100)]
Fix function that shouldn't be globally defined.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agoTrivial coding style fix.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:29:52 +0000 (11:29 +0100)]
Trivial coding style fix.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agoAdd function declarations, and ensure we prevent mismatch
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:28:12 +0000 (11:28 +0100)]
Add function declarations, and ensure we prevent mismatch
of function definition vs. declaration.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agoReplace 0 with NULL when assigning pointer fields/parameters.
kaf24@firebug.cl.cam.ac.uk [Wed, 16 Nov 2005 10:27:19 +0000 (11:27 +0100)]
Replace 0 with NULL when assigning pointer fields/parameters.

Signed-off-by: Vincent Hanquez <vincent@xensource.com>
20 years agoFix 64-bit build that I broke by making acpi table mapping
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 18:46:48 +0000 (19:46 +0100)]
Fix 64-bit build that I broke by making acpi table mapping
fixup dependent on CONFIG_XEN_X86 (that only xen/i386
defines).

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoMake watchStart and watchMain global functions rather than classmethods,
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 18:19:02 +0000 (19:19 +0100)]
Make watchStart and watchMain global functions rather than classmethods,
meaning that we no longer need to prefix all the field accesses with cls.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 18:09:30 +0000 (19:09 +0100)]
Merged.

20 years agoHave a watch callback return 0 or 1 depending upon whether it would like to
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 18:08:11 +0000 (19:08 +0100)]
Have a watch callback return 0 or 1 depending upon whether it would like to
continue to receive watches.  This means that it is not necessary to probe
around in the xswatch internals to unregister a watch.

Tidy up the hotplug watch handling, moving the nested function out to a
separate place (I don't think that this was a problem in the end, but it was
making me nervous as I was debugging the recent race condition, and I find it
clearer out of the class in any case.  Pass an integer code representing
hotplug status, once it has been parsed from the store, as there are now a few
different places we can detect failure, and it's cleaner to pass a code from
those places.

Small tidy up to XendDomain, removing the unused releaseDomain field.

20 years agoRemove the handling of TypeError inside gather. It is not thrown by int()
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 17:47:39 +0000 (18:47 +0100)]
Remove the handling of TypeError inside gather.  It is not thrown by int()
when given a non-convertible value, as claimed by the comment, as the exception
ValueError is thrown in that situation.  TypeError is only thrown on int(None),
which cannot actually take place in this code, because of the explicit check
for None.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoEnsure we initialise the cpu_present_map before
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 17:43:28 +0000 (18:43 +0100)]
Ensure we initialise the cpu_present_map before
topology_init() is called. In latest Linux kernels it
iterates over cpu_present_map rather than cpu_possible_map.

Also, save/restore should really iterate over
possible cpus, not present ones (not that it really matters,
but it's a small cleanup).

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoThis patch fixes dom0 SMP vcpu hotplug. Currently, domains without
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 16:30:55 +0000 (17:30 +0100)]
This patch fixes dom0 SMP vcpu hotplug.  Currently, domains without
config files (e.g. dom0) don't set info['vcpus'] correctly resulting in
incorrect cpu availablity values in the store.  This prevents hotplug
from functioning.

Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
20 years agoThis patch fixes a number of vcpu related issues.
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 16:28:05 +0000 (17:28 +0100)]
This patch fixes a number of vcpu related issues.

1.  xm vcpu-list now shows info for all vcpus by using
    info['max_vcpu_id']+1 as end of the range to query
    vcpu_info.
2.  Add new vcpu fields, online_vcpus, max_vcpu_id to
    XendDomainInfo.sxpr()
3.  Remove filter_cpump() which isn't needed now that
    the per vcpu cpumap field is correct.
4.  Update xm/main.py to use online_vcpus as the number
    vcpus to display in list.

Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 16:22:42 +0000 (17:22 +0100)]
Merged.

20 years agoFix race between xspy_read_watch and xspy_watch, by placing the watch in the
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 16:22:04 +0000 (17:22 +0100)]
Fix race between xspy_read_watch and xspy_watch, by placing the watch in the
xshandle's list of watches *before* registering the watch with Xend.

Closes bug #392.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoacpi change is Xen/x86 specific, so make it dependent on
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 16:02:37 +0000 (17:02 +0100)]
acpi change is Xen/x86 specific, so make it dependent on
CONFIG_XEN_X86.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoUpgrade Xen's msr.h to include definitions from Linux 2.6.14.
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 15:57:02 +0000 (16:57 +0100)]
Upgrade Xen's msr.h to include definitions from Linux 2.6.14.
Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 15:24:31 +0000 (16:24 +0100)]
Merged.

20 years agoReduce the time required to start xm-test domains by allowing the console
danms@us.ibm.com [Tue, 15 Nov 2005 15:07:44 +0000 (16:07 +0100)]
Reduce the time required to start xm-test domains by allowing the console
to retry connecting to the DomU multiple times.  A polling approach instead
of a wait-and-hope-it-works approach.

20 years agoHave the vif scripts expect the offline event, not the remove event. With
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 14:56:47 +0000 (15:56 +0100)]
Have the vif scripts expect the offline event, not the remove event.  With
the remove event, the script environment variable has disappeared, so we must
hook on the offline event instead.  This fix was put into xen-backend.rules,
but not the vif scripts, so they were out of sync, and the interfaces were not
being properly closed down.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoAdded link_exists function, for neatness. Check for the existence of the pdev
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 14:54:51 +0000 (15:54 +0100)]
Added link_exists function, for neatness.  Check for the existence of the pdev
before issuing the diagnostic about the missing vdev.  Having a pdev but no
vdev is a strong indicator that the network-bridge script has run once already,
and so there is nothing to worry about.  Having the vdev and no corresponding
pdev is still diagnosed.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 14:44:37 +0000 (15:44 +0100)]
Merged.

20 years agoThis patch is largely stylistic, and gets rid of StuddlyCaps formatting in
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 14:36:42 +0000 (15:36 +0100)]
This patch is largely stylistic, and gets rid of StuddlyCaps formatting in
all the param variables, make consistant use of bold and italics in the the
man pages to make it easier to read, and adds a few minor content additions.
It makes actually reading the man page a lot easier.

Signed-off-by: Sean Dague <sean@dague.net>
20 years agoThis patch adds more content around the scheduler portions of the xm.1 man
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 14:27:48 +0000 (15:27 +0100)]
This patch adds more content around the scheduler portions of the xm.1 man
page.  Most of this content came from the xensource wiki or the sedf doc in
the Xen tree.  It still needs more work, but it is a good first start.

Signed-off-by: Sean Dague <sean@dague.net>
20 years agoThis patch is another in the continued effort to flesh out the man pages, it
emellor@leeni.uk.xensource.com [Tue, 15 Nov 2005 14:26:37 +0000 (15:26 +0100)]
This patch is another in the continued effort to flesh out the man pages, it
adds pretty reasonable text to most of the code xen domain and host commands
(device and scheduler will come in the future).

Signed-off-by: Sean Dague <sean@dague.net>
20 years agoUpdate hgignore list.
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 14:11:24 +0000 (15:11 +0100)]
Update hgignore list.

20 years agoThe new userland monitoring tool, XenMon.
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 14:09:58 +0000 (15:09 +0100)]
The new userland monitoring tool, XenMon.
Signed-off-by: Rob Gardner <rob.gardner@hp.com>
20 years agoExtra tracing in Xen to support XenMon.
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 13:59:59 +0000 (14:59 +0100)]
Extra tracing in Xen to support XenMon.
Signed-off-by: Rob Gardner <rob.gardner@hp.com>
I also appied some fixes and simplifications to the tracing.
Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoPrevious changeset trivially broke ioemu/helper2.c. Fix it.
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 10:36:44 +0000 (11:36 +0100)]
Previous changeset trivially broke ioemu/helper2.c. Fix it.
Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoThis patch fixed keyboard issue for linux kernel 2.6 in vmx guest.
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 10:34:29 +0000 (11:34 +0100)]
This patch fixed keyboard issue for linux kernel 2.6 in vmx guest.
linux 2.6 kernel keyboard use a timer to poll the kbd state,in
virtualization environment,the kbd interrupt injection is very likely
to happend just after kbd state query in the kbd timer and thus
will reverse sequence of scan code.
Also fix env->send_event SMP issue.

Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com>
Signed-off-by: Edwin Zhai<edwin.zhai@intel.com>
Signed-off-by: Eddion Dong <eddie.dong@intel.com>
Signed-off-by: Asit Mallick <asit.k.mallick@intel.com>
20 years agoProperly declare ioapic_update_EOI().
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 10:28:33 +0000 (11:28 +0100)]
Properly declare ioapic_update_EOI().

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoInitialise secondary CPUs via an arch_initcall rather than
kaf24@firebug.cl.cam.ac.uk [Tue, 15 Nov 2005 10:28:01 +0000 (11:28 +0100)]
Initialise secondary CPUs via an arch_initcall rather than
a subsys_initcall. This ensures that initialisation occurs
before topology_init().

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoget_page_type() must clean up writable pagetable state
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 18:55:40 +0000 (19:55 +0100)]
get_page_type() must clean up writable pagetable state
before failing, otherwise functions like set_gdt() can
spuriously fail because a page appears to be writable
despite there being pending changes to mapping of that
page in ptwr state.

This should fix the reported cases of crashes in
vcpu_prepare in xenlinux.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoBetter fix for flushing conflicting batched ptwr state
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 18:45:32 +0000 (19:45 +0100)]
Better fix for flushing conflicting batched ptwr state
before emulating a PT access.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoFlush writable pagetable state before emulating a PT
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 17:27:11 +0000 (18:27 +0100)]
Flush writable pagetable state before emulating a PT
update. Avoids possibility of updating a PTE temporarily
marked writable by ptwr batching logic, which can corrupt
page reference counts. Aiee!

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoUse make_lowmem_page_readonly/writable() in preference to the
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 17:13:38 +0000 (18:13 +0100)]
Use make_lowmem_page_readonly/writable() in preference to the
generic functions where appropriate. This prevents us using the
generic functions early during boot, when pte_pfn() does not work
(because max_mapnr is not initialised).

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoFailure to revalidate a writable pagetable page should crash
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 14:21:16 +0000 (15:21 +0100)]
Failure to revalidate a writable pagetable page should crash
the offending domain, not merely pause it. Also, callers can
now specify another domain to domain_crash().

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoFix check for a working iptables installation. Checking for the iptables
emellor@leeni.uk.xensource.com [Mon, 14 Nov 2005 11:28:47 +0000 (12:28 +0100)]
Fix check for a working iptables installation.  Checking for the iptables
binary is not sufficient, because the user may not have the appropriate
modules installed.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Mon, 14 Nov 2005 11:12:29 +0000 (12:12 +0100)]
Merged.

20 years agoAdded line of debugging to try and catch hotplug script race.
emellor@leeni.uk.xensource.com [Mon, 14 Nov 2005 11:11:41 +0000 (12:11 +0100)]
Added line of debugging to try and catch hotplug script race.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoDeclare local variables at the start of a block, not in the middle
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 10:47:38 +0000 (11:47 +0100)]
Declare local variables at the start of a block, not in the middle
(ISO C constraint).

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoClean up xenstore/console shared page initialisation, which is
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 10:36:42 +0000 (11:36 +0100)]
Clean up xenstore/console shared page initialisation, which is
now handled solely by the domain builder.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoRevert accidentally applied changes from changeset
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 10:35:50 +0000 (11:35 +0100)]
Revert accidentally applied changes from changeset
7783:5aad7e145e501fbfb346954950a33b042a963633.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoTrivial Makefile cleanup for ioemu.
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 10:11:11 +0000 (11:11 +0100)]
Trivial Makefile cleanup for ioemu.

20 years agoAllocate MAC addresses using Xensource Inc's new OUI.
emellor@leeni.uk.xensource.com [Mon, 14 Nov 2005 10:06:25 +0000 (11:06 +0100)]
Allocate MAC addresses using Xensource Inc's new OUI.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoIf /sbin/ isn't in the path, udev rules will erroneously not get
kaf24@firebug.cl.cam.ac.uk [Mon, 14 Nov 2005 10:05:34 +0000 (11:05 +0100)]
If /sbin/ isn't in the path, udev rules will erroneously not get
installed.

Signed-off-by: Nivedita Singhvi <niv@us.ibm.com>
20 years agoThe various changes to move to using info instead of passing the config
emellor@leeni.uk.xensource.com [Sat, 12 Nov 2005 17:59:56 +0000 (18:59 +0100)]
The various changes to move to using info instead of passing the config
sxpr everywhere broke the use of a boot loader for booting with.  The
attached fixes things up so that
a) use of a bootloader ==> kernel not specified in the config file
b) don't cleanup bootloader files until after the domain has booted
c) rerun bootloader on reboot

Signed-off-by: Jeremy Katz <katzj@redhat.com>
20 years agoFurther changes to xm.1 man page.
emellor@leeni.uk.xensource.com [Sat, 12 Nov 2005 17:52:16 +0000 (18:52 +0100)]
Further changes to xm.1 man page.

Signed-off-by: Sean Dague <sean@dague.net>
20 years agoFrom Charles Coffing <ccoffing@novell.com>:
emellor@leeni.uk.xensource.com [Sat, 12 Nov 2005 17:41:37 +0000 (18:41 +0100)]
From Charles Coffing <ccoffing@novell.com>:

"make uninstall" was not cleaning up some udev rules.  Also, on x86_64
there was stuff left in /usr/lib/xen, not just /usr/$LIBDIR/xen.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoThe following patch is work in progress to add more meat to the xm.1 man
emellor@leeni.uk.xensource.com [Sat, 12 Nov 2005 17:32:19 +0000 (18:32 +0100)]
The following patch is work in progress to add more meat to the xm.1 man
page.  Expect more content to trickle in over time.

Signed-off-by: Sean Dague <sean@dague.net>
20 years agoThis test was failing for he wrong reason. Calls to block-attach after the
dan@guaranine.beaverton.ibm.com [Sat, 12 Nov 2005 17:25:12 +0000 (18:25 +0100)]
This test was failing for he wrong reason.  Calls to block-attach after the
first *should* report failure, so the test should not fail because of this.
The test now fails because multiple calls to block-attach seem to detach the
block devicem, which is wrong.

Signed-off-by: Dan Smith <danms@us.ibm.com>
20 years agoDisable use of mmap. This allows Xenstored to run with an NFS-mounted
emellor@leeni.uk.xensource.com [Sat, 12 Nov 2005 16:51:14 +0000 (17:51 +0100)]
Disable use of mmap.  This allows Xenstored to run with an NFS-mounted
/var/lib/xenstored.  Closes bug #400.  This fix no doubt has a performance hit,
so I have opened bug #401 to discuss a better fix.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 20:59:33 +0000 (21:59 +0100)]
Merged.

20 years agoNew Xenbus driver structure, bringing an explicit state machine to the table.
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 20:59:05 +0000 (21:59 +0100)]
New Xenbus driver structure, bringing an explicit state machine to the table.
This will make the drivers more robust to failure of the other end, and makes
current driver state more visible.  It also allows us to bring very similar
pieces of code from the drivers together into the xenbus driver.  This should
be the start of a fix to the problem with block and net device hotplugging,
because the code was previously confused about the difference between the
connection state of the kernel device layer, and the state of the xenbus
communication.  This confusion is addressed here.

Much of this work is influenced by Rusty Russell's skeleton driver, recently
posted to xen-devel, as that piece made suggestions as to structure and
arrangement that have been taken on board, albeit pushed down into the xenbus
layer.

A new file xenbus_client.c has been added.  This file implements the interface
between the xenbus layer and the layer specific to the driver.  Access to
lower layers is still allowed for those who do not need all the facilities of
the new driver.

xenbus_dev_error has moved from xenbus_xs.c into xenbus_client.c, and
xenbus_dev_ok has been removed, in favour of xenbus_switch_state.

DevController.py now closes down a device by writing a change to the Closing
state on the backend, to allow the frontend time to close down properly.  Its
also writes the frontend and backend details within the same transaction.

The timeout on the hotplug scripts has been cranked down from 2 minutes to 5
seconds.

The diagnose script has been updated to parse the new states.

The blktab and tpm drivers have yet to be ported to this structure.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoAttached patch allows PAE xenlinux to run in the shadow mode using
smh22@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 18:11:13 +0000 (19:11 +0100)]
Attached patch allows PAE xenlinux to run in the shadow mode using
log-dirty guest-refcount, which is required to support
save/restore/relocate. We can turn on/off the mode every 5-sec interval
while doing kernel build (make -j4), for example, and it survives for
hours.

We are still restoring log-dirty mode for x86_64 xenlinux, which is not
very stable right now, but I believe it should be done very soon.

We also checked that it did not break 64-bit VMX domains, which uses
different sub-mode of the shadow mode.

Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>
Signed-off-by: Xiaohui Xin <xiaohui.xin@intel.com>
20 years agoFix make_page_readonly/make_page_writeable on PAE guests - previous behaviour
smh22@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 18:02:49 +0000 (19:02 +0100)]
Fix make_page_readonly/make_page_writeable on PAE guests - previous behaviour
was incorrect when machine address > 4GB. We believe this fixes bugzilla #267.

Also tidy up and fix show_page_walk.

Signed-off-by: Steven Hand <steven@xensource.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoRename the commands inside the hotplug scripts to match those used by hotplug
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 17:01:44 +0000 (18:01 +0100)]
Rename the commands inside the hotplug scripts to match those used by hotplug
and udev.  Pass all commands through from xen-backend.agent or
xen-backend.rules rather than second-guessing the individual scripts.  This
should make it easier to tear down devices cleanly, especially the dangling
vifs we are seeing at the moment.

Cope with a missing type node inside the block script, as we don't want to
resurrect a removed node by writing a hotplug error message.

Log failure of iptables command.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoNew location for code common to both netfront and netback, with the MAC parsing
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 16:59:20 +0000 (17:59 +0100)]
New location for code common to both netfront and netback, with the MAC parsing
pulled here.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoDon't call iptables if it is not installed. This code is only there to ensure
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 16:56:02 +0000 (17:56 +0100)]
Don't call iptables if it is not installed.  This code is only there to ensure
that packets between domU and the bridge are not firewalled, so if iptables is
not installed, then there is no worry.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoFix function call syntax in vif-route and vif-nat.
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 16:04:17 +0000 (17:04 +0100)]
Fix function call syntax in vif-route and vif-nat.

Function calls in shell script should not contain parenthesis.

Signed-off-by: Murillo Fernandes Bernardes <mfb@br.ibm.com>
20 years agoMerged.
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 14:36:16 +0000 (15:36 +0100)]
Merged.

20 years agoChmod a+x.
emellor@leeni.uk.xensource.com [Fri, 11 Nov 2005 14:36:07 +0000 (15:36 +0100)]
Chmod a+x.

20 years agoMissed two files from NULL-pointer-check changeset.
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 13:54:20 +0000 (14:54 +0100)]
Missed two files from NULL-pointer-check changeset.
Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoRemove unnecessary NULL checks before freeing memory blocks.
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 11:55:47 +0000 (12:55 +0100)]
Remove unnecessary NULL checks before freeing memory blocks.
The following functions check for NULL internally:
 free, kfree, xfree, free_xenheap_page[s]

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoFix application of OS patches to deal better with failure.
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 10:16:46 +0000 (11:16 +0100)]
Fix application of OS patches to deal better with failure.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoMax xen skbuff allocation is now 64KB. kmem caches only
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 10:01:32 +0000 (11:01 +0100)]
Max xen skbuff allocation is now 64KB. kmem caches only
get populated with objects on first allocation, so if large
caches end up being unused, they don't actually cost us
very much.

Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoMerge with xen-unstable-ia64.hg
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 09:47:43 +0000 (10:47 +0100)]
Merge with xen-unstable-ia64.hg

20 years agonloopbacks default is now 8. So vifnum of greater than 7 requires
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 09:46:36 +0000 (10:46 +0100)]
nloopbacks default is now 8. So vifnum of greater than 7 requires
an adjustment to nloopbacks. Warning comment updated.

Signed-off-by: Nivedita Singhvi (niv@us.ibm.com)
20 years agoAdd MOVSX instr support to VMX MMIO decoder.
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 09:41:25 +0000 (10:41 +0100)]
Add MOVSX instr support to VMX MMIO decoder.

Signed-off-by: Xin Li <xin.b.li@intel.com>
20 years agoRemove unnecessary ifdef in defn of xprintk. Install
kaf24@firebug.cl.cam.ac.uk [Fri, 11 Nov 2005 09:39:20 +0000 (10:39 +0100)]
Remove unnecessary ifdef in defn of xprintk. Install
xenguest.h.
Signed-off-by: Joseph Cihula <joseph.cihula@intel.com>
20 years agoEnable gcc3.4 with CONFIG_SMP (by Alex Williamson)
djm@kirby.fc.hp.com [Thu, 10 Nov 2005 21:24:29 +0000 (15:24 -0600)]
Enable gcc3.4 with CONFIG_SMP (by Alex Williamson)

20 years agoBlkfront no longer built by default, need CONFIG_XEN_BLKDEV_FRONTEND
djm@kirby.fc.hp.com [Thu, 10 Nov 2005 20:02:01 +0000 (14:02 -0600)]
Blkfront no longer built by default, need CONFIG_XEN_BLKDEV_FRONTEND

20 years agoFix Linux fault.c indentation.
kaf24@firebug.cl.cam.ac.uk [Thu, 10 Nov 2005 18:43:56 +0000 (19:43 +0100)]
Fix Linux fault.c indentation.
Signed-off-by: Keir Fraser <keir@xensource.com>
20 years agoFix page-table dump on PAE with >4GB MFNs.
smh22@firebug.cl.cam.ac.uk [Thu, 10 Nov 2005 18:32:09 +0000 (19:32 +0100)]
Fix page-table dump on PAE with >4GB MFNs.

Signed-off-by: Steven Hand <steven@xensource.com>
20 years agoMerge with http://xenbits.xensource.com/ext/xen-unstable-docs.hg
kaf24@firebug.cl.cam.ac.uk [Thu, 10 Nov 2005 17:25:27 +0000 (18:25 +0100)]
Merge ... xenbits.xensource.com/ext/xen-unstable-docs.hg

20 years agoMerge
djm@kirby.fc.hp.com [Thu, 10 Nov 2005 16:51:37 +0000 (10:51 -0600)]
Merge

20 years agoAdd "Securing Xen" adapted from Anthony Liguori's Wiki entry.
Robb Romans [Thu, 10 Nov 2005 16:43:24 +0000 (11:43 -0500)]
Add "Securing Xen" adapted from Anthony Liguori's Wiki entry.

20 years agoA few doc cleanups. Replace bitkeeper with mercurial.
Robb Romans [Thu, 10 Nov 2005 16:42:58 +0000 (11:42 -0500)]
A few doc cleanups. Replace bitkeeper with mercurial.

20 years agoReflect various missing traps/faults (flagged with printf for now).
djm@kirby.fc.hp.com [Thu, 10 Nov 2005 16:39:38 +0000 (10:39 -0600)]
Reflect various missing traps/faults (flagged with printf for now).
Without this building xen documentation crashes Xen!
Signed-off by: Dan Magenheimer <dan.magenheimer@hp.com>

20 years agoThis patch use same expression as vif for vmx guest but has a distinguish "type=ioemu"
Xiaofeng Ling [Thu, 10 Nov 2005 16:07:13 +0000 (17:07 +0100)]
This patch use same expression as vif for vmx guest but has a distinguish "type=ioemu"
just like block device.
e.g. vif=['type=ioemu, mac=01:00:00:00:00:33, bridge=xenbr0']
type=ioemu to specify for vmx NIC device.
nics=4 can specify 4 vmx NIC now.
bridge is passed to device model now.
With this patch, the error
"Error: Device 0 (vif) could not be connected."
will not happen for creating vmx guest
currently, it happens on some linux distribution.

Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com>
20 years agoRemoved autogenerated file.
emellor@leeni.uk.xensource.com [Thu, 10 Nov 2005 15:59:51 +0000 (16:59 +0100)]
Removed autogenerated file.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoFix the log reopening by moving the code with all the races out of the signal
emellor@leeni.uk.xensource.com [Thu, 10 Nov 2005 15:20:27 +0000 (16:20 +0100)]
Fix the log reopening by moving the code with all the races out of the signal
handler and into the main loop.  The "self-pipe trick" is used to awaken the
select() call on SIGHUP.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
20 years agoFix typo
djm@kirby.fc.hp.com [Thu, 10 Nov 2005 15:07:47 +0000 (09:07 -0600)]
Fix typo

20 years agoMerge latest xen-unstable into xen-ia64-unstable
djm@kirby.fc.hp.com [Thu, 10 Nov 2005 14:34:09 +0000 (08:34 -0600)]
Merge latest xen-unstable into xen-ia64-unstable