Commit Graph
6107 Commits
Author SHA1 Message Date
kaf24@firebug.cl.cam.ac.uk 19b77ba87c [ We'll try to get in a proper fix into 3.0 - this is just the backup plan]
Don't attempt to create paravirtualized devices for VMX domains for now.

Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:05:58 +00:00
kaf24@firebug.cl.cam.ac.uk 094f89109d Fix comments about the localtime option.
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:05:21 +00:00
kaf24@firebug.cl.cam.ac.uk eaba06e1d3 Clean up unused junk in xemexample.vmx
Signed-off-by: YongKang You <yongkang.you@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:04:35 +00:00
kaf24@firebug.cl.cam.ac.uk 66c1bb81a1 Need to reset device model pid on reboot
Without this, xend will not spawn a new device model process after the reboot.

Signed-off-by: Ke Yu <ke.yu@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:04:17 +00:00
kaf24@firebug.cl.cam.ac.uk ca21e61a00 Support rebooting VMX domains
Signed-off-by: Ke Yu <ke.yu@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:03:16 +00:00
kaf24@firebug.cl.cam.ac.uk f1e7438c35 Handle VMX domains correctly across xend restarts
This patch remembers the device model pid across xend restarts and
avoids creating duplicate device model processes.

Also, device models don't inherit socket fds from xend (which are closed
on exec now), which used to prevent xend restarting.

Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: Xiaohui Xin <xiaohui.xin@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:02:51 +00:00
kaf24@firebug.cl.cam.ac.uk 55aaa284cf Separate out VMCS host state initialization from the rest
This is needed to support virtual cpu migration, because the host state
initialization is done on every migration, but the rest of the initialization
is done once per vcpu.

Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:01:36 +00:00
kaf24@firebug.cl.cam.ac.uk cd5f4dd4c3 Improved error reporting on vmlaunch/vmresume failure.
Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 21:01:10 +00:00
kaf24@firebug.cl.cam.ac.uk ab6d52907c Fix config file parsing for VMX domains.
If we define "vif" in the config file, image.py will raise error "vmx:
missing vbd configuration". The reason is "vif" is dealt with as a "vbd"
device.

This patch fixes this issue by dealing with "vbd" and "vif " separately,
removing "macaddr" arg and parsing mac address from "vif" instead.

Also, the vbd doesn't have to be a file anymore, but can be a physical
disk partition.

Signed-off-by: Yunfeng Zhao <yunfeng.zhao@intel.com>
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
2005-08-11 20:58:08 +00:00
kaf24@firebug.cl.cam.ac.uk 2521f14bda This patch addresses Rustys' [1]suggestions. Most visibly:
Entries in the store:
/cpus/cpuX/online -> /cpu/X/availability

Values:
0|1 -> online|offline

Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
2005-08-11 20:57:09 +00:00
kaf24@firebug.cl.cam.ac.uk c4e885cb53 Fix handling of memory holes for Xen heap and domain 0
kernel image and ramdisk.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-11 18:03:22 +00:00
kaf24@firebug.cl.cam.ac.uk a606f89f0e Clean up event-channel notification code in Xen.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-11 16:48:36 +00:00
kaf24@firebug.cl.cam.ac.uk 4afea39c56 Fix where "!" operator used in Bitwise operation.
In IBM we have an internal source code scanner called BEAM. We have run
it against Xen and here are some of the results.

Signed-off-by: Jerone Young <jyoung5@us.ibm.com>
2005-08-11 16:23:54 +00:00
kaf24@firebug.cl.cam.ac.uk e20872cf5e More time fixes. Now time is set properly in domU's, and
settimeofday() is simplified and fixed.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-11 13:07:08 +00:00
twilkie@paploo.uk.xensource.com 3140b13b50 Added support for abbreviated xm sub commands 2005-08-11 12:06:32 +00:00
kaf24@firebug.cl.cam.ac.uk 0b3823d582 PAE page directories must be below 4GB. Based on a patch
from Gerd Knorr.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-11 10:34:32 +00:00
kaf24@firebug.cl.cam.ac.uk 954a35eeb5 The following patch provides x86_64 SMP support for xen linux. Many,
many thanks to the help for bug fixes, cleanups, good domU support,
and more from Jun and Xin.  I've tested this on a couple different
machines, and made sure the UP build still boots and works.  dom0 and
domU smp both work.  There's still some cleanups to do, but we'd prefer
to do the last bit in tree.  This doesn't update defconfigs just yet.
That'll be a follow-on patch shortly, so for now, you'll have to manually
enable CONFIG_SMP if you'd like to test.

Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Xin Li <xin.b.li@intel.com>
Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>
2005-08-11 08:59:47 +00:00
kaf24@firebug.cl.cam.ac.uk 5dba835ba5 Since Xend doesn't use twisted anymore, remove the references from the
Makefile.

Signed-off-by: Mark Williamson <mark.williamson@cl.cam.ac.uk>
2005-08-11 08:58:40 +00:00
kaf24@firebug.cl.cam.ac.uk 66e693a5c0 Remove qemu-vgaram-bin. It's not used any more. 2005-08-11 08:57:48 +00:00
kaf24@firebug.cl.cam.ac.uk 2f403612d8 The attached patch fixes the xm list command, which very
recently stopped working if security was enabled.

xm list fails if security is enabled because of a single
additional "%" character in a print.

Signed-off-by: Reiner Sailer <sailer@watson.ibm.com>
2005-08-11 08:55:47 +00:00
cl349@firebug.cl.cam.ac.uk 3a264694de merge? 2005-08-10 21:55:36 +00:00
iap10@firebug.cl.cam.ac.uk 7b4525c5aa Enable reiserfs in x86_64 domU kernel
Signed-off-by: ian@xensource.com
2005-08-10 17:48:03 +00:00
cl349@firebug.cl.cam.ac.uk b34f79ff69 merge? 2005-08-10 17:06:29 +00:00
iap10@firebug.cl.cam.ac.uk 90c0e3301d Install libxenstore in /usr/lib64 for 64 bit builds
Signed-off-by: ian@xensource.com
2005-08-10 16:52:03 +00:00
cl349@firebug.cl.cam.ac.uk 282447bac6 Make xenconsole more friendly when invoked directly.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-08-10 16:47:07 +00:00
kaf24@firebug.cl.cam.ac.uk 1c57a9e116 Small time fixes.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-10 16:03:31 +00:00
kaf24@firebug.cl.cam.ac.uk b6e5a7b02b Fix wallclock initialisation in domain0.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-10 15:14:50 +00:00
kaf24@firebug.cl.cam.ac.uk b1231e1b08 Fix domain0 builder for PAE.
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-10 14:11:40 +00:00
kaf24@firebug.cl.cam.ac.uk add4888a4d Fix 64-bit __get_user() on 32-bit x86. Needed for PAE.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-10 13:06:00 +00:00
kaf24@firebug.cl.cam.ac.uk 2de676fedb Fix stupid typo in previous checkin (stopped time.c
from compiling on 32-bit x86).
2005-08-10 13:05:22 +00:00
kaf24@firebug.cl.cam.ac.uk e9041e1e82 Fix time handling on x86/64.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-10 12:56:47 +00:00
twilkie@paploo.uk.xensource.com cbc34f7812 Fix simple index error in error handling. Now reports correct domain name in error message. 2005-08-10 12:21:39 +00:00
kaf24@firebug.cl.cam.ac.uk 509bb24d17 merge 2005-08-10 09:50:15 +00:00
kaf24@firebug.cl.cam.ac.uk 15f4fae9dc dma_map_single can't sleep since it may be called with locks held, or
interrupts off, etc.   Thus it shouldn't make sleeping allocations.

Signed-off-by: Chris Wright <chrisw@osdl.org>
2005-08-10 09:50:05 +00:00
cl349@firebug.cl.cam.ac.uk efd6f7db4d merge? 2005-08-10 09:47:13 +00:00
cl349@firebug.cl.cam.ac.uk c2bf763b7f Make xen daemon startup more robust.
1) make xenconsoled exit gracefully if xcs dies
2) daemonize xenstored before binding to xenstored socket
3) wait to close stdio in xenstored until we're ready to accept
connections (so that PID=`xenstored --output-pid` doesn't return until
it's ready to accept connections)
4) updates tools/misc/xend for these changes
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustycorp.com.au>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-08-10 09:46:51 +00:00
kaf24@firebug.cl.cam.ac.uk a4857a1c3f Add padding to allocation bitmap. Should fix page fault
on x86/64 debug builds on some systems.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-08-10 09:45:06 +00:00
kaf24@firebug.cl.cam.ac.uk 42c9a99912 The following patch adds cramfs support to domU kernels so cramfs based
ramdisks can be used for initrds.

Signed-off-by: Sean Dague <sean@dague.net>
2005-08-10 09:09:15 +00:00
cl349@firebug.cl.cam.ac.uk 1b8d0b8824 Initialize the xenstore MFN for VMX domains
Signed-off-by: Arun Sharma <arun.sharma@intel.com>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-08-10 08:44:58 +00:00
cl349@firebug.cl.cam.ac.uk 9ae7cd4f23 merge? 2005-08-09 16:26:19 +00:00
cl349@firebug.cl.cam.ac.uk 9045816f69 Do not suppress errors on writes outside transactions.
Mike Wray made this change because xend's callbacks were failing
with EROFS, because a transaction was going on.  However, the
solution is to either block callbacks during transactions, or use
a different connection to the store.
We fail ex-transaction activity because it's not clear what the right
behaviour is: it allows deadlock if two people do it, and it implied
your transaction isn't rooted in the right place.

This hack caused the testsuite to fail (obviously), so as a result the
testsuite wasn't run very often.  This caused it to bitrot.
Signed-off-by: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-08-09 16:16:08 +00:00
kaf24@firebug.cl.cam.ac.uk 191a7d28f8 Directly including ../tools breaks building in separate object tree,
as well as building from mkpatch patch (I believe Rik hit that one).
This puts xenstore headers in linux-public.

Signed-off-by: Chris Wright <chrisw@osdl.org>
2005-08-09 13:53:15 +00:00
kaf24@firebug.cl.cam.ac.uk 09d42caafd Add missing LGPL license notice to python sources
that didn't have them.

Signed-off-by: Mike Wray <mike.wray@hp.com>
2005-08-09 13:17:37 +00:00
kaf24@firebug.cl.cam.ac.uk ba529a426c PAE PMDs must be made writable before freeing. 2005-08-09 13:08:25 +00:00
kaf24@firebug.cl.cam.ac.uk e7f2cf60b5 merge 2005-08-09 13:04:54 +00:00
kaf24@firebug.cl.cam.ac.uk cd6fd4d17c merge 2005-08-09 13:04:31 +00:00
twilkie@paploo.uk.xensource.com 8aa46fd245 Updates to SV. New look. Removed and simplified lots of files. 2005-08-09 12:55:18 +00:00
cl349@firebug.cl.cam.ac.uk 670f10866e merge? 2005-08-09 12:51:06 +00:00
cl349@firebug.cl.cam.ac.uk b9a5e01c19 The following patch is a major restructuring of main.py to be more straight forward, and remove as much code as possible.
It does a number of things:

* It implements the standardized options proposed via my RFC for xm
* It restructures the calling of these functions to be function pointers out
  of a dictionary, instead of subclasses found through inspection
* It consolidates all help info, making it easier to see that the help
  screen looks consistent across commands.  (Per command help isn't yet in
  this patch just global help)
* It moves object importing into the functions that require them only,
  allowing commands like "xm help" to be runable as non root
* It adds command aliasing to provide backwards compatibility with many
  commands (i.e. xm balloon)
* It provides more useful error messages is most fail cases
* It implements a short help by default (which takes < 24 screen lines) as
  requested by Mark W at OLS
* It attempts to trap more exceptions than before and provide useful errors
  based on them (this requires more testing)

This patch makes no changes to any other elements of xen/xm, and works with
all other objects and libraries as is.

Signed-off-by: Sean Dague <sean@dague.net>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-08-09 12:50:22 +00:00
kaf24@firebug.cl.cam.ac.uk 1787e1a4b7 When building in separate object directory entry.o fails building
because entry.S includes syscall_table.S but it's in the wrong
directory.  This simply adds the obj-dir to the include search path.

Signed-off-by: Chris Wright <chrisw@osdl.org>
2005-08-09 10:47:41 +00:00