Commit Graph
6083 Commits
Author SHA1 Message Date
kaf24@firebug.cl.cam.ac.uk ecfbf99a10 This patch renames balloon command to "set-mem", so instead of "xm
balloon", we will use "xm set-mem" (as suggested by TODO list)

Signed-off-by: Nguyen Anh Quynh <aquynh@gmail.com>
2005-07-29 10:37:30 +00:00
kaf24@firebug.cl.cam.ac.uk 084dedfee3 This patch does 2 jobs:
- Enforce the number of CPUs dom0 will take. See the new variable
"dom0-cpus" in xend-config.sxp (you will want to set this variable to
1 on SMP systems)

- Balloon out dom0 memory when creating domU, if there is not enough
free memory. The lowest level we will balloon out is configured via
the new variable "dom0-min-mem" in xend-config.sxp


I still have a doubt: where to put the code to enforce dom0-cpus. At
the moment I put it into
python/xen/xend/server/SrvDaemon.py, and hopefully that is resonable
enough. Any comment?



Signed-off-by: Nguyen Anh Quynh <aquynh@gmail.com>
2005-07-29 10:36:53 +00:00
kaf24@firebug.cl.cam.ac.uk 613f9362cf Patch for 64-bit VMX guest destroy
64-bit VMX guest can't be destroyed without this patch, as guest page's
reference count problem.
This patch made two point of reference count correct:
1. In invlpg, the old page count need put.
2. There is a specail case for fl1's shadow reference count.=20
=20

Signed-off-by: Chengyuan Li <chengyuan.li@intel.com>
Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>
2005-07-29 10:36:11 +00:00
kaf24@firebug.cl.cam.ac.uk 12b1a321bd Here is our latest patch to enable the Xen tools to
build and work with IA64 Xen.
Signed-off-by: Matt Chapman <matthewc@hp.com>
Acked-by: Dan Magenheimer <dan.magenheimer@hp.com>
2005-07-29 10:34:45 +00:00
kaf24@firebug.cl.cam.ac.uk 8a9db38089 Attached patch adds a DMA zone to xen, also modifies xen_contig_memory()
to ask for DMA pages.
Signed-off-by: srparish@us.ibm.com
2005-07-29 10:31:22 +00:00
kaf24@firebug.cl.cam.ac.uk 4e6f46cf10 This patch updates the CPU field in xm list for multi-vcpu domains.
When a dom runs on multiple physical cpus, the CPU field doesn't have
much meaning.  The suggested fix was to replace the value with '-'.  
Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
2005-07-29 10:27:12 +00:00
kaf24@firebug.cl.cam.ac.uk 259b555e84 ... for dom0 if we know we'll need it. This is the xen
counterpart to the previous xenlinux patch which adds this
info to the kernel.  Also adds some printk's.
2005-07-29 10:24:41 +00:00
kaf24@firebug.cl.cam.ac.uk 654ee452ae ... so it builds again and also add a note to the xen elf header
that this kernel should be booted in shadow-translated mode.
2005-07-29 10:24:06 +00:00
kaf24@firebug.cl.cam.ac.uk 1c5c1e73ac Really just basic preparation: switch over PAE builds to the new
shadow code, drop old dummy functions, add (fewer) new ones.
2005-07-29 10:23:07 +00:00
kaf24@firebug.cl.cam.ac.uk c71983a0a5 fix x86_64 domain0 /dev/mem issue
x86_64 miss the define of ARCH_HAS_DEV_MEM, that causes
/drivers/char/mem.c to not show low 1M memory correctly.

Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com>
2005-07-29 10:22:03 +00:00
ach61@arcadians.cl.cam.ac.uk a087d98570 PDB: add some additional files 2005-07-28 21:32:13 +00:00
ach61@arcadians.cl.cam.ac.uk 476167d4df PDB: process targets 2005-07-28 21:28:23 +00:00
fred@localhost.localdomain 6e3fd6a74d Make ns16550 rework on tiger4 box
Signed-off-by Kevin Tian <kevin.tian@intel.com>
2005-07-28 07:33:34 -08:00
fred@localhost.localdomain 31fdbf01d9 Add kregs change 2005-07-28 07:29:12 -08:00
fred@localhost.localdomain e05e9a82ed This patch adds necessary support for vcontext merge between
para and vti side. Per Dan's good suggestion, all the changes
are refined into ia64 specific. Accompanied with xeno patches
and a bit libxc change, xen0 + xenU can still work again.

Signed-off-by Eddie Dong <eddie.dong@intel.com>
Signed-off-by Kevin Tian <kevin.tian@intel.com>
Signed-off-by Fred Yang <fred.yang@intel.com>
2005-07-28 07:16:12 -08:00
cl349@firebug.cl.cam.ac.uk fcb76453f3 merge? 2005-07-28 12:34:45 +00:00
cl349@firebug.cl.cam.ac.uk 0c4fa371ca Don't unbind port when event channel binding fails.
This happens when xend tries to introduce the same domain twice.
Signed-off-by: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-28 12:33:53 +00:00
cl349@firebug.cl.cam.ac.uk 107f383ae1 Simplify code by acking watch as soon as we read it.
Since watches no longer block anyone, this doesn't change semantics.
Signed-off-by: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-28 12:29:23 +00:00
cl349@firebug.cl.cam.ac.uk 8b3e983883 Simplify code by making in and out pointers internal to xenbus_xs.c
Signed-off-by: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-28 12:25:13 +00:00
kaf24@firebug.cl.cam.ac.uk cbc208facb Nicer platform timer string (include clock rate in mhz).
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-07-28 10:20:08 +00:00
fred@localhost.localdomain 2875cb7f90 More syntactic changes for preparation of vcpu_info merge
Signed-off-by Kevin Tian <Kevin.Tian@intel.com>
2005-07-28 01:48:15 -08:00
kaf24@firebug.cl.cam.ac.uk 678f2597f6 Fix irq.c warning, from aq. 2005-07-28 09:37:53 +00:00
fred@localhost.localdomain 63ebfaf017 Add emulation to rsm.be, which is first encounted in guest fast syscall
path, where little endian is forced just after epc.

Signed-off-by Kevin Tian <kevin.tian@intel.com>
2005-07-28 01:26:30 -08:00
kaf24@firebug.cl.cam.ac.uk e4dbaf3593 Fix gnttab bug in blkfront.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-07-28 09:10:05 +00:00
fred@localhost.localdomain a570f1c3ac Fix the issue of complete rebuild for any change
Signed-off-by Kevin Tian <kevin.tian@intel.com>
2005-07-28 00:57:06 -08:00
fred@localhost.localdomain 6dfdc5d0cd Nop change. Just a test 2005-07-27 23:08:20 -08:00
kaf24@firebug.cl.cam.ac.uk dddab7e3c4 Slightly better time debugging support in Xen. 2005-07-27 23:15:15 +00:00
djm@kirby.fc.hp.com 5c2e42c0b8 Disable NaT support, fast tick and improved PAL support until
multi-domain stabilizes
2005-07-27 13:16:28 -07:00
djm@kirby.fc.hp.com bb9c51809d Syntactic changes to prepare for possible vcpu_info changes 2005-07-27 12:57:05 -07:00
djm@kirby.fc.hp.com 4c2b7ffd2c Add xen-shared-info offset generation
Signed-off-by: Eddie Dong <eddie.dong@intel.com>
Signed-off-by: Kevin Tien <kevin.tien@intel.com>
Signed-off-by: Fred Yang <fred.yang@intel.com>
2005-07-27 12:07:07 -07:00
cl349@firebug.cl.cam.ac.uk 1b268774e8 Get rid of internet_copy.jpg once more.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 18:41:49 +00:00
cl349@firebug.cl.cam.ac.uk 14302840ff Don't ignore exceptions here - the possible RuntimeError is handled already.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 18:40:32 +00:00
cl349@firebug.cl.cam.ac.uk 272942eeb3 Introduce dom0 to the store.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 18:31:18 +00:00
cl349@firebug.cl.cam.ac.uk 001ee4ddde Handle introduce_domain failing when the domain is already introduced.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 18:31:03 +00:00
cl349@firebug.cl.cam.ac.uk 7aa5afeb6c Always pass around the store machine frame number instead of the linear address.
Also cleanup IOCTL_PRIVCMD_INITDOMAIN_STORE.
Signed-off-by: Rusty Russel <rusty@rustcorp.com.au>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 17:17:09 +00:00
cl349@firebug.cl.cam.ac.uk f2cbf339d4 Add xc_init_store python binding.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 17:06:02 +00:00
cl349@firebug.cl.cam.ac.uk 57103c6152 Fix read of integers from database on event channel restore.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 17:04:47 +00:00
cl349@firebug.cl.cam.ac.uk 55e1cda4c0 Make XENSTORED_DEBUG create a xenstored trace logfile.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 17:03:00 +00:00
cl349@firebug.cl.cam.ac.uk 0e358737de Move SIF_BLK_BE_DOMAIN and SIF_NET_BE_DOMAIN definitions to where they are used.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 17:00:24 +00:00
cl349@firebug.cl.cam.ac.uk f423cd3752 merge? 2005-07-27 16:57:37 +00:00
cl349@firebug.cl.cam.ac.uk 63c82e1bdd Don't auto-add domains we don't know about.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
2005-07-27 16:57:01 +00:00
kaf24@firebug.cl.cam.ac.uk 8ab01c502c Add IBM Cyclone support to Xen.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-07-27 16:18:02 +00:00
kaf24@firebug.cl.cam.ac.uk 5e8bc03f67 Add HPET support to Xen. It is quicker to access and more
precise than the PIT.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-07-27 15:32:44 +00:00
kaf24@firebug.cl.cam.ac.uk e386da97d1 On opt_nosmp, also force num_cores to 1. Do this in setup.c
instead of cpu/common.c. This is where opt_nosmp is defined,
and it reduces diff of cpu/ subdir against native Linux.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-07-27 11:20:47 +00:00
sos22@douglas.cl.cam.ac.uk b5deb01bfa Kill a bad assertion.
Signed-off-by: Steven Smith, sos22@cam.ac.uk
2005-07-27 11:02:30 +00:00
sos22@douglas.cl.cam.ac.uk 1a92d96790 Merge.
Signed-off-by: Steven Smith, sos22@cam.ac.uk
2005-07-27 10:54:39 +00:00
sos22@douglas.cl.cam.ac.uk fb6e1a7ed1 Make sure that opt_nosmp also disables hyperthreading. This stops us
from trying to create domains on non-existent threads, which would
lead to a rather nasty crash.

Signed-off-by: Steven Smith, sos22@cam.ac.uk
2005-07-27 10:52:29 +00:00
cl349@firebug.cl.cam.ac.uk 1693f2a523 Rename file with space in name.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>


--HG--
rename : tools/sv/images/internet copy.jpg => tools/sv/images/internet_copy.jpg
2005-07-27 09:56:31 +00:00
kaf24@firebug.cl.cam.ac.uk 6e27226763 Fix memory reservation in XenLinux.
Signed-off-by: Keir Fraser <keir@xensource.com>
2005-07-27 09:24:51 +00:00
iap10@freefall.cl.cam.ac.uk 36df52b0a8 Frob the Linux-2.6 configs to enable PAE (HIGHMEM_64G) if XEN_TARGET_X86_PAE=y
Signed-off-by: ian@xensource.com
2005-07-27 00:27:36 +00:00