smh22@firebug.cl.cam.ac.uk [Wed, 7 Dec 2005 17:59:33 +0000 (18:59 +0100)]
Updated info on Xend.
Signed-off-by: Nivedita Singhvi <niv@us.ibm.com>
Signed-off-by: Robb Romans <FMJ@us.ibm.com>
Signed-off-by: Steven Hand <steven@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 16:26:07 +0000 (16:26 +0000)]
Don't try and print out the info returned by the server after a restore -- there
isn't any. Don't try and unpause the restored domain either -- XendCheckpoint
does that.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 16:25:13 +0000 (16:25 +0000)]
Use xm create's --dryrun option to get the name from the configuration file,
rather than the eversoskanky approach taken previously (use grep, and replace
any occurrences of %d with [0-9]* !)
Signed-off-by: Ewan Mellor <ewan@xensource.com>
danms@us.ibm.com [Wed, 7 Dec 2005 15:50:59 +0000 (15:50 +0000)]
Make the xm-test ramdisk makefile correct the permissions on the rcS script
Signed-off-by: Dan Smith <danms@us.ibm.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 15:44:38 +0000 (15:44 +0000)]
Fix the startup behaviour when the tdb file exists. We were attempting to
remove and recreate the /local directory, but the manual_node call only works
if the tdb is empty -- we must use create_node, otherwise the store is
corrupted.
It's not clear to me that the other manual_node calls are safe, but I have only
made this smaller change for the sake of stability. These other calls should
be reviewed.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 15:41:41 +0000 (15:41 +0000)]
Fix usleep definition -- we would be sleeping 1000 times too long if the usleep
command were not available elsewhere on the machine (on my Debian boxes, for
example).
Sleep for a configurable amount of time between xm create calls (by default,
5 seconds). This spaces out the booting of the new domains, meaning that they
should not thrash the disk so much.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 11:57:26 +0000 (11:57 +0000)]
Tidy up use of environment variables in xenbus-hotplug interaction.
I've changed this to only write "backend" (xenbus_backend.root) into
XENBUS_BASE_PATH. After appending XENBUS_TYPE (also exported) you can
construct the "base path" easily in the scripts. I've completely
removed writing XENBUS_FRONTEND_ID in favour of reading it from the
store at "$XENBUS_PATH/frontend-id".
Avoid re-reading the store over and over for the frontend domain's vm-path.
The patch also fixes a memory leak in backend_bus_id where we
leak the memory referenced by frontend if the function succeeds.
This work is by Christian Limpach.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 11:51:31 +0000 (11:51 +0000)]
Merged.
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 11:50:55 +0000 (11:50 +0000)]
Change configuration handling wrt to the vcpus entry, so that we only see a
vcpus entry in the image section when using VMX, only add VMX options at all
if the builder is vmx, and issue a warning when overriding the global vcpus
setting with the one from the image.
This removes the duplicated vcpus entries in the domain's sxpr.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 11:47:46 +0000 (11:47 +0000)]
Remove timeout on select -- there is no need for it, since we don't do anything
in the main loop except service I/O.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 11:44:02 +0000 (11:44 +0000)]
Merged.
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 11:43:43 +0000 (11:43 +0000)]
Further fix to same_vm check. This fixes intermittent failure of live
localhost migration.
return [ test ] is not valid, simply [ test ] at the end of the function is the
correct syntax.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 10:54:57 +0000 (10:54 +0000)]
Fix printf of ssize_t that broke 64-bit build.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 10:53:47 +0000 (10:53 +0000)]
Squelch error messages from calls when failure is handled gracefully.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 10:51:55 +0000 (10:51 +0000)]
Remove one of the multiple occurrences of vcpus in the sxpr. This value is
in ROUNDTRIPPING_CONFIG_ENTRIES, so there is no need for it to be mentioned
explicitly in sxpr() too.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Wed, 7 Dec 2005 10:27:03 +0000 (10:27 +0000)]
xm-test report fails because IA64's /cpu/procinfo format is different from
x86's. Here's a small patch to fix the problem.
Signed-off-by: Matsumoto <n_matumoto@soft.fujitsu.com>
emellor@leeni.uk.xensource.com [Tue, 6 Dec 2005 17:11:04 +0000 (17:11 +0000)]
Merged.
emellor@leeni.uk.xensource.com [Tue, 6 Dec 2005 17:10:59 +0000 (17:10 +0000)]
Add a "consumed" parameter to the buffer declaration, and use this to advance
through the buffer when the data are read by console clients. This avoids a
costly memmove with every read from the console.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 16:48:57 +0000 (17:48 +0100)]
Pull nmi/traps changes from Linux 2.6.14.
Signed-off-by: Keir Fraser <keir@xensource.com>
emellor@leeni.uk.xensource.com [Tue, 6 Dec 2005 16:40:50 +0000 (16:40 +0000)]
Merged.
emellor@leeni.uk.xensource.com [Tue, 6 Dec 2005 16:40:43 +0000 (16:40 +0000)]
Make the master fd belonging to the pty pair non-blocking. Writes were blocking
if the console client disconnected before all the data had been written to the
pty, even though the select call had indicated that the fd was previously
writable. Reads have been observed exhibiting similar behaviour too.
Closes bug #380.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 15:05:31 +0000 (16:05 +0100)]
The Mini-OS build once again broke, this time since vcpu_data became vcpu_info
and that XEN_VER in the __xen_guest section changed slightly. I've modified
the mini-os so that it at least compiles and boots again.
Signed-off-by: Simon Kagstrom <simon.kagstrom@bth.se>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 14:52:51 +0000 (15:52 +0100)]
Add map_domain_pages/unmap_domain_pages to map contiguous
multi-franme extents of domain memory.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 14:52:15 +0000 (15:52 +0100)]
Fix the request size for a full page-sized skbuff in netfront
driver (must allow 16 bytes for dev_alloc_skb headroom).
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 14:02:54 +0000 (15:02 +0100)]
Fix netfront to accept received packets at a wider range of
offsets within a memory page. Also fix, skb realloc-and-copy
when copying really is required (we were forgetting whether
the packet data is known valid, and checksumming packets that
do not have the csum field filled in).
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 11:49:05 +0000 (12:49 +0100)]
Pre-decrement rather than post-decrement in shadow32.c.
Already fixed in shadow.c.
Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 11:46:43 +0000 (12:46 +0100)]
Update xen cap to reflect 32 bit VMX guest support on x86_64
Signed-off-by: Xin Li <xin.b.li@intel.com>
kaf24@firebug.cl.cam.ac.uk [Tue, 6 Dec 2005 11:45:30 +0000 (12:45 +0100)]
Fix vtpm support in xend.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
dan@guaranine.beaverton.ibm.com [Mon, 5 Dec 2005 15:36:11 +0000 (15:36 +0000)]
Output an explicit newline before the "REASON:" string so the report script
can find it, even when a domain outputs a non-newline-terminated line.
Signed-off-by: Dan Smith <danms@us.ibm.com>
emellor@leeni.uk.xensource.com [Mon, 5 Dec 2005 15:30:10 +0000 (15:30 +0000)]
Mention udev, clean up xm help section, mention network-nat and network-route
scripts and the fact that network is now called network-bridge.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
iap10@freefall.cl.cam.ac.uk [Mon, 5 Dec 2005 13:02:01 +0000 (14:02 +0100)]
smh22@firebug.cl.cam.ac.uk [Mon, 5 Dec 2005 12:39:26 +0000 (13:39 +0100)]
Updated docs for Xen 3.0.
Signed-off-by: Steven Hand <steven@xensource.com>
cl349@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 19:12:00 +0000 (20:12 +0100)]
Temporarily remove NetBSD and FreeBSD sparse trees to avoid user confusion.
The NetBSD and FreeBSD currently don't build against the final Xen 3.0 API.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
cl349@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 17:24:24 +0000 (18:24 +0100)]
Merge
kaf24@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 17:20:41 +0000 (18:20 +0100)]
Pull requests from blkfront into blkback a bit more
aggressively, to keep the scheduling pipeline filled.
Signed-off-by: Keir Fraser <keir@xensource.com>
cl349@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 17:17:33 +0000 (18:17 +0100)]
Update version number for Xen 3.0.0 release.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
cl349@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 17:16:33 +0000 (18:16 +0100)]
Make version strings consistent across all interfaces.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
cl349@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 17:12:18 +0000 (18:12 +0100)]
Update -xen config file to include newgen megaraid and qla scsi drivers.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
cl349@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 16:29:46 +0000 (17:29 +0100)]
Rename set-vcpus to vcpu-set (match docs) and update short help page.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
kaf24@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 11:06:38 +0000 (12:06 +0100)]
Bump blktap and dom0op version numbers.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 10:57:32 +0000 (11:57 +0100)]
Image.py should send 2 new arguments to qemu-dm for setting QEMU window
Title. 1 is const string "-domain-name", the other is the string of
domain name. At present imagy.py combines these two strings to 1. So
Qemu will fail to recognize it and fail to start.
Signed-off-by: Yongkang You <yongkang.you@intel.com>
kaf24@firebug.cl.cam.ac.uk [Sun, 4 Dec 2005 10:54:04 +0000 (11:54 +0100)]
Add a comment to tricky corner case in evtchn_close().
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Keir Fraser <keir@xensource.com>
rread@ubuntu.eng.hq.xensource.com [Sun, 4 Dec 2005 00:52:38 +0000 (17:52 -0700)]
Merge docs
iap10@freefall.cl.cam.ac.uk [Sun, 4 Dec 2005 00:34:25 +0000 (01:34 +0100)]
Readme typos
Signed-off-by: ian@xensource.com
cl349@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 23:51:15 +0000 (00:51 +0100)]
Fix same_vm test by adding missing '$'.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
cl349@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 23:49:51 +0000 (00:49 +0100)]
Use test ([) instead of expr to compare integers.
expr $int1 '<' $int2 seems to be broken on some shells.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 20:53:01 +0000 (21:53 +0100)]
Fix request-notification holdoff in blkback. New code is
more correct and clearer in intent.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 20:17:11 +0000 (21:17 +0100)]
Remove scary warning message about 'breaking irq affinity'
when hot-unplugging cpus. It is a harmless side-effect of
running the irqbalance daemon in user space.
Signed-off-by: Keir Fraser <keir@xensource.com>
iap10@freefall.cl.cam.ac.uk [Sat, 3 Dec 2005 19:16:04 +0000 (20:16 +0100)]
Merge
iap10@freefall.cl.cam.ac.uk [Sat, 3 Dec 2005 19:05:39 +0000 (20:05 +0100)]
Update xend-config.sxp so that dom0-min-mem is 196MB by default, as this is rather more user friendly.
Also do some minor fixes to the README
Signed-off-by: ian@xensource.com
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 18:09:19 +0000 (19:09 +0100)]
Always initialise a swiotlb for domain0. Make it a small
aperture (2MB) for small systems (< 2GB).
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 16:59:37 +0000 (17:59 +0100)]
Enable multiple NICs for vmx domain by configuration file
Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 16:58:31 +0000 (17:58 +0100)]
Set the device model windows title as the domain name
This helps when creating multile vmx domain.
Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 16:41:16 +0000 (17:41 +0100)]
Fixes unmodified 32-bit guest support on the x86_64 Xen.
Signed-off-by: Jun Nakajima <jun.nakajima@intel.com>
Signed-off-by: Xiaohui Xin <xiaohui.xin@intel.com>
cl349@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 12:36:22 +0000 (13:36 +0100)]
Fix block backend teardown when the device didn't setup completely.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
emellor@leeni.uk.xensource.com [Sat, 3 Dec 2005 12:21:52 +0000 (12:21 +0000)]
Merged.
emellor@leeni.uk.xensource.com [Sat, 3 Dec 2005 12:21:27 +0000 (12:21 +0000)]
Set hostname in DHCP config from the guest name, and if more than one interface
is specified, from guest-2, guest-3 etc. Have the DHCP server send the
hostname to the guest.
Added a lock around DHCP config file frobbing, to avoid racing creation vs
deletion.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 10:41:49 +0000 (11:41 +0100)]
Fix blkback request notification holdoff. req_cons must be
updated before make_response() is called. This fixes attaching
ramdisk-backed vbds to a blkfront.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 10:00:03 +0000 (11:00 +0100)]
Fix typo in previous changeset. Include is ../grant_table.h
not ../gnttab.h.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 09:50:22 +0000 (10:50 +0100)]
Inclide grant_ref_t definition in split-driver struct
defns.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 09:44:38 +0000 (10:44 +0100)]
Merge with xen-ia64-unstable.hg
kaf24@firebug.cl.cam.ac.uk [Sat, 3 Dec 2005 09:44:01 +0000 (10:44 +0100)]
Using a slow serial console with sync_console on SMP leaves an
unusable system: it gets into a flood of
Timer ISR/0: Time went backwards: delta=-
3566279 cpu_delta=
16433721
shadow=
2019998373 off=
420435384 processed=
2444000000
cpu_processed=
2424000000
0:
2424000000
1:
2444000000
messages which swamps the console and leaves things almost, but not
quite, completely hung. It is a simple matter to ratelimit these
printks, which completely eliminates the problem.
kmself@ix.netcom.com [Sat, 3 Dec 2005 02:59:53 +0000 (19:59 -0700)]
Incorporating Alan Oehler's changes, some of mine.
emellor@leeni.uk.xensource.com [Sat, 3 Dec 2005 00:51:36 +0000 (00:51 +0000)]
Read from the store if the filename reported by losetup has been truncated (the
loop driver stores only the first 64 chars of the filename). In this case, we
need to read through the store until we find the domain using the device, and
read the configured filename from there instead.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kmself@ix.netcom.com [Sat, 3 Dec 2005 00:13:55 +0000 (17:13 -0700)]
Incorporating Alan's mods.
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 22:27:04 +0000 (22:27 +0000)]
Merged.
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 21:52:41 +0000 (15:52 -0600)]
Revert to simpler ia64 defconfigs, generic one causes loopback failures
kmself@ix.netcom.com [Fri, 2 Dec 2005 21:50:08 +0000 (14:50 -0700)]
Applied Robb Roman's patches to upstream.
Robb Romans [Fri, 2 Dec 2005 21:29:26 +0000 (14:29 -0700)]
Small cleanup for distro chapters.
Signed-off-by: Robb Romans <FMJ@us.ibm.com>
Robb Romans [Fri, 2 Dec 2005 21:29:26 +0000 (14:29 -0700)]
Reorg patch 2 to match wiki.xensource.com/xenwiki/Xen3DocsToDo
Second patch to reorganize the manual to match the structure in the
Xen3DocsToDo Wiki entry.
Robb Romans [Fri, 2 Dec 2005 21:29:25 +0000 (14:29 -0700)]
Reorg patch 1 to match wiki.xensource.com/xenwiki/Xen3DocsToDo
First patch to reorganize the manual to match the structure in the
Xen3DocsToDo Wiki entry.
Incorporates a patch from Nivedita Singhvi to clean up the Further
Support Chapter.
Signed-off-by: Robb Romans <FMJ@us.ibm.com>
cl349@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 21:15:58 +0000 (22:15 +0100)]
Fix vif-net script for setups where eth0 has no IP address.
Use the IP address associated with the backend network interface
for connections from dom0 to the guest connected to this network
interface.
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 18:55:00 +0000 (12:55 -0600)]
Minor changes due to file movement to merge xen-unstable with xen-ia64-unstable
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 18:52:25 +0000 (12:52 -0600)]
Merge
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 18:26:58 +0000 (19:26 +0100)]
Fix PAE bootstrap.
Signed-off-by: Keir Fraser <keir@xensource.com>
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 18:12:11 +0000 (12:12 -0600)]
Stack pal call emulation implemented. (by Tristan Gingold)
PAL_HALT emulation added: stop machine if dom0, shutdown domain otherwise.
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 17:56:24 +0000 (18:56 +0100)]
Add a defensive batched tlb flush to free_page_type(), to
ensure the linear_pg_table remains in sync with modified
page table structure. Otherwise we can update stale entries
and screw reference counts (but probably only when running
a malicious domain).
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 17:00:49 +0000 (18:00 +0100)]
Move cpu0_stack out of Xen text section and into BSS. This
avoids getting loads of bogus cpu0_stack lines in call
backtraces from non-debug builds.
Doing this requires greater alignment of the BSS section,
which reveals a bug in ld where the alignment padding is
not added to the program segment's memsz field. We get around
this by finding the address of the last symbol in the image,
and increasing our load image's memsz to include that symbol.
Also some cleanups to the linker scripts.
Signed-off-by: Keir Fraser <keir@xensource.com>
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 16:27:22 +0000 (10:27 -0600)]
Update xen-capable elilo to support compressed dom0 kernel images (by Alex Williamson)
This will be released in next official version of elilo but is included
in Xen/ia64 for developer and user convenience until the new version of
elilo is more widely distributed.
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 16:12:07 +0000 (10:12 -0600)]
Enable serial input for Xen (tied to ifdef, default off for now)
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
djm@kirby.fc.hp.com [Fri, 2 Dec 2005 16:06:42 +0000 (10:06 -0600)]
Fix pointer in vmx vmmu init code (by Anthony Xu and Kevin Tian)
dan@guaranine.beaverton.ibm.com [Fri, 2 Dec 2005 15:57:24 +0000 (15:57 +0000)]
Make "xm list --long" generate device configuration data from xenstore state
instead of internal xend state.
Signed-off-by: Dan Smith <danms@us.ibm.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:52:47 +0000 (15:52 +0000)]
This patch adds a new domain config option, 'cpus' which is a list of
CPUs a domains' vcpus can use. The older 'cpu' config option is
prepended to the list of cpus to use and will keep the behavior of
pinning VCPU0.
The cpus option supports ranges and negation, so:
cpus = "0-3,5,^1" produces -> [0,2,3,5]
The list is circular, so in a domain with the following config:
vcpus = 4
cpus = "0,3,7" # Use any of 0, 3, 7 for this domain.
would see vcpus 0-3 pinned to cpus 0,3,7,0 respectively.
Also, the pin operation is moved before the memory reservation as vcpu
to cpu mapping will be helpful for future NUMA work when trying to
allocate pages close to the physical cpus being used.
An update to the display of cpumap was needed to normalize the cpumap
values to the range of possible cpus.
I've also included some text for the xmdomain.cfg(5) man page.
Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:48:55 +0000 (15:48 +0000)]
Merged.
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:48:44 +0000 (15:48 +0000)]
Fix the block-sharing check for physical devices by using a lock to serialise
the check, reading from the store rather than /sys, and checking whether the
VM for apparently-conflicting uses is actually the same (i.e. this is a
migration or a reboot in progress).
This fixes recent failures in 12_block_attach_shared_domU and recloses bug #331.
This fix includes particularly skanky path slicing inside xenbus_probe, which
it would be nice to remove very soon.
The loopback device check still suffers from the problem that filenames over
64 characters long are truncated.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:39:13 +0000 (15:39 +0000)]
Added /etc/init.d/dhcpd to list of init files to check for (supporting Centos
4.2).
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:38:22 +0000 (15:38 +0000)]
Added serialisation facilities. Udev is supposed to serialise events, but sadly
the older hotplug infrastructure does not, so we must be able to serialise the
block-sharing checks (etc) ourselves.
Redirect stderr to /var/log/xen-hotplug.log, for better debugging.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:36:39 +0000 (15:36 +0000)]
Remove the watch on the /vm/<uuid> path when rebooting or migrating. We don't
want to end up with multiple watches on the same path, otherwise we get a
watch event storm after many such reboots or migrates.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:35:22 +0000 (15:35 +0000)]
Don't log every time around the big transaction writing the device details.
This transaction may be starved for a short while, and logging inside the while
loop won't help!
Use new xstransact.complete to tidy up one other transaction.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
emellor@leeni.uk.xensource.com [Fri, 2 Dec 2005 15:33:38 +0000 (15:33 +0000)]
Added xen-hotplug.log to the files to collect.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 15:26:57 +0000 (16:26 +0100)]
More console-related cleanups.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 15:14:20 +0000 (16:14 +0100)]
Simplify console driver and avoid unnecessary spinning in the
transmit handler.
Signed-off-by: Keir Fraser <keir@xensource.com>
iap10@freefall.cl.cam.ac.uk [Fri, 2 Dec 2005 14:37:31 +0000 (15:37 +0100)]
Disable devfs from the -xen configs. It's obsolete and was causing problems for some users.
Signed-off-by: ian@xensource.com
sos22@douglas.cl.cam.ac.uk [Fri, 2 Dec 2005 14:16:37 +0000 (15:16 +0100)]
Avoid a hang when probing the partition table on imported block
devices. If you export a device so the importing domain sees it as a
whole disk rather than a partition, then add_disk will try to probe
its partition table. However, we were calling add_disk before the
device was properly connected, and then not connecting it until
add_disk had finished. This meant that we ended up never actually
connecting the device, and any accesses to it would then hang.
Fix this by not calling add_disk until we're connected.
Signed-off-by: Steven Smith, sos22@cam.ac.uk
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 12:27:39 +0000 (13:27 +0100)]
Blkif request notifications use generic ring req_event holdoff
mechanism rather than custom mechanism.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 12:16:13 +0000 (13:16 +0100)]
Add padding to ring headers for future expansion.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 12:10:25 +0000 (13:10 +0100)]
Clean up netif messages. Remove bitfields, and create
extensible flag fields in tx requests and rx responses, in
place of specific csum_blank/csum_valid bitfields.
Signed-off-by: Keir Fraser <keir@xensource.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 10:10:01 +0000 (11:10 +0100)]
Fix tpm split driver to include grant-table defintions.
Signed-off-by: Stefan Beger <stefanb@us.ibm.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 10:08:24 +0000 (11:08 +0100)]
While working bug #143, it came to my attention that no verbose error is
loged if the dom0 kernel mangles a packet and tries to checksum
offload it.
This should let the user know what is going wrong (instead of silently
dropping the packet).
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
kaf24@firebug.cl.cam.ac.uk [Fri, 2 Dec 2005 10:06:40 +0000 (11:06 +0100)]
Fix __get_user() asm warning by making source and dest
operands the same size (32 bits).
Signed-off-by: Keir Fraser <keir@xensource.com>