proxmox-mirrors/qemu-server
1
0
Fork 0
mirror of https://git.proxmox.com/git/qemu-server synced 2025-05-08 21:05:23 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
762 Commits 7 Branches 0 Tags 12 MiB
2ed5d5724c
Commit Graph

762 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Dietmar Maurer
be19058305 bump version to 3.0-22 
and delete trailing white spaces
 2013-07-15 09:16:01 +02:00
Alexandre Derumier
462e8d19fe add hyper-v enlightments for windows guests 
This add special hyper-v cpu flags for windows guests.

This improve performance and avoid some bsod related to timer.

(I currently disable the hv_vapic flag because I can't get it working).

I have tested all theses flags with: win2003, win2008R2, winxp, linux debian 64bit, on intel and amd physicals processor

It doesn't break live migration, because new cpu flags are not see by guests until a vm reset.

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-07-15 09:09:08 +02:00
Alexandre Derumier
2e1a5389f7 add +sep cpuflags to kvm32 && kvm64 cpu models 
Need for win8 boot.

This flag was missing from rhel < 6.4 host kernel. It's ok now.
But it's also missing from kvm64 model. (It's exist in other cpu models, amd or intel).
So it's pretty safe to enable it.
If the host kernel is older, qemu filter the flag.

This also improve performance of winxp && win7 32 bits guests.

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-07-15 09:08:50 +02:00
Alexandre Derumier
519ed28c96 enable x2apic by default for kvm machines 
This reduce context switch with multicore guests.

Even if the host cpu don't have x2apic, it's working because qemu have an virtual x2apic implementation for guest.
We need  in-kernel irqchip support for this, which is enable for kvm guest since qemu 1.3.
(I don't enable it if nokvm param is set)

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-07-15 09:08:34 +02:00
Dietmar Maurer
716a470c8c spiceproxy: set delete-this-file 2013-06-27 09:28:13 +02:00
Alexandre Derumier
af32683a67 spiceconfig: port can't be 0 
Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-06-27 06:12:46 +02:00
Alexandre Derumier
e554e5b9a9 spice : fix proxy address 
Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-06-27 06:10:34 +02:00
Dietmar Maurer
46246f04e9 vm status API: return info about spice 2013-06-26 13:37:14 +02:00
Dietmar Maurer
013d8cacc6 bump version to 3.0-21 2013-06-26 13:16:19 +02:00
Dietmar Maurer
3309e65afa spiceproxy: remove socat, and return data to access the new spiceproxy server 2013-06-26 13:13:58 +02:00
Dietmar Maurer
eb15737004 kill socat if we do not get any connection within $timeout seconds 2013-06-25 14:34:59 +02:00
Dietmar Maurer
d914737577 depend on socat (used for spice) 2013-06-25 13:14:00 +02:00
Dietmar Maurer
8bd0cbf5c3 spice proxy: try to cirrectly count open socat connections 2013-06-25 12:54:15 +02:00
Dietmar Maurer
5ecf258f2c limit spice ticket length to 59 charachters 2013-06-25 12:47:33 +02:00
Dietmar Maurer
288eeea8ae add spice proxy API 
This is experimental code, spice connections are not encryped and thus insecure.
We use ticket passwords for spice auth, and do direct spice connections to
the nodes instead of using a tunnel.
 2013-06-25 12:09:05 +02:00
Dietmar Maurer
1011b57090 add support for spice display driver (vga=qxl) 
This is based on the patch from Alexandre, but we use vga=qlx instead
of introducing a new 'spice: 1' boolean flag.
 2013-06-25 07:10:42 +02:00
Dietmar Maurer
9b2c0efbd6 correctly handle undef values when calling qemu_block_set_io_throttle 2013-06-12 07:08:02 +02:00
Alexandre Derumier
0e6165343a qemu-server : bridge_add : fix uninitialized value 
fix : Use of uninitialized value $bridgeid in numeric lt (<) at /usr/share/perl5/PVE/QemuServer.pm line 2774.

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-06-12 06:58:20 +02:00
Dietmar Maurer
7043d946fc code cleanups 
removed trailing white space
 2013-06-11 07:27:31 +02:00
Dietmar Maurer
d52b8b77f8 simplify previous patch 2013-06-11 07:22:13 +02:00
Alexandre Derumier
09a8989501 create_disks : avoid storage scan v3 
Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-06-11 07:09:17 +02:00
Dietmar Maurer
89b0baa650 bump version to 3.0-20 2013-06-07 11:44:42 +02:00
Dietmar Maurer
5555edea95 update_vm_async: new asynchronous API 2013-06-07 11:41:58 +02:00
Dietmar Maurer
d8b916fdb0 snapshot rollback: use pc-i440fx-1.4 as default 2013-06-06 06:38:12 +02:00
Dietmar Maurer
6dbe8b459b require VM.Config.HWType priviledges to modify 'machine' configuration 2013-06-06 06:22:32 +02:00
Dietmar Maurer
bf6d72aa13 bump version to 3.0-19 2013-06-05 10:28:08 +02:00
Dietmar Maurer
42668529e6 migrate: pass --machine parameter to remote 'qm start' command 2013-06-05 10:24:39 +02:00
Dietmar Maurer
4b15803dee snapshot: save/restore 'machine' configuration 2013-06-05 10:17:49 +02:00
Dietmar Maurer
ff556cf2a0 implement get_current_qemu_machine 2013-06-05 09:52:22 +02:00
Dietmar Maurer
952958bc87 API extension: add 'machine' parameter to vm_start 
This allows us to overwrite the machine type at start - useful for
migration to make sure we use the same machine at the target node.
 2013-06-05 09:27:31 +02:00
Dietmar Maurer
3bafc5106e add new 'machine' configuration 2013-06-05 09:18:23 +02:00
Dietmar Maurer
1a71fa73a4 fix typo 2013-06-04 13:33:35 +02:00
Dietmar Maurer
2fc6bc1722 qemu_drive_mirror: correctly call block-job-cancel if block-job-complete fails 2013-05-31 11:31:22 +02:00
Dietmar Maurer
37eccdb5bc bump version to 3.0-18 2013-05-31 10:57:59 +02:00
Dietmar Maurer
70d45e33e9 move_disk: implement delete flag 2013-05-31 10:56:16 +02:00
Dietmar Maurer
43bc02a921 rename move to move_disk 
Because that API call can only move a single disk (not the whole VM).
 2013-05-31 08:46:21 +02:00
Dietmar Maurer
7baa079ca5 bump version to 3.0-17 2013-05-29 13:01:45 +02:00
Dietmar Maurer
1377d7b0f5 clone_disk: test if requested format is supported - else use default 2013-05-29 12:59:21 +02:00
Dietmar Maurer
635c3c4426 move_vm: add original disk as unusedX 2013-05-29 12:15:30 +02:00
Dietmar Maurer
e2cd75fa65 move_vm: make it work with PVE::QemuServer:::clone_disk 
- use POST instead of PUT (this is not idempotent).
- return task ID
- use declarative permission check
 2013-05-29 12:07:56 +02:00
Alexandre Derumier
586bfa78d1 add qm move (storage migration) 
qm move <vmid> <disk> <storage> --format [raw|qcow2|vmdk]

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-05-29 11:23:45 +02:00
Dietmar Maurer
c55fee03a0 allow to clone VM when name is not set 2013-05-29 11:18:54 +02:00
Dietmar Maurer
152fe752c4 factor out clone_disk 
As suggested by Alexandre. Also cleanup  qemu_drive_mirror (fix white space).
 2013-05-29 08:32:10 +02:00
Dietmar Maurer
a8c7e31e60 bump version to 3.0-16 2013-05-28 12:24:13 +02:00
Dietmar Maurer
05937a147d qm rescan: do not add aliases as unused disks. 2013-05-28 12:21:14 +02:00
Dietmar Maurer
a8e2f942bd fix unused disk handling 
Show unused disks even if disk is used inside snapshots. But do not allow
to remove those disks.
 2013-05-28 12:08:45 +02:00
Dietmar Maurer
7a907ce612 rescan: add check if config has really changed 2013-05-27 10:51:01 +02:00
Dietmar Maurer
5996a936e3 fix bug 395: correctly handle unused disk with storage alias 2013-05-27 08:25:39 +02:00
Dietmar Maurer
13471a97fb bump version to 3.0-15 2013-05-23 11:34:09 +02:00
Alexandre Derumier
e9cfd9cb01 qemu_img_format : use raw for as default for other storage (nexenta, sheepdog, ...) 
Seem that something have change in qemu.

iscsi qemu block driver need to use raw, or we get a segfault. (like rbd).

sheepdog works also with .raw, so we don't need anymore specific storage format, we can use raw.

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
 2013-05-23 11:33:17 +02:00