-by: Stefan Reiter
---
Whoops wrong file, sorry for noise.
v3 -> v4:
* use correct test description for new test
v2 -> v3:
* include test case and fix existing simple1 test
v1 -> v2:
* handle error in kvm_user_version
* corrected commit message (2.12 -> 2.8)
PVE/QemuServer.pm
-by: Stefan Reiter
---
v2 -> v3:
* include test case and fix existing simple1 test
v1 -> v2:
* handle error in kvm_user_version
* corrected commit message (2.12 -> 2.8)
PVE/QemuServer.pm | 14 +-
test/cfg2cmd/old-qemu.conf | 4
test/cfg2cmd/simple1.conf |
in the report mentions), this series
would indeed allow you to specify it as a custom CPU model instead of
having to use the '-args' parameter.
Hope that helps,
Stefan
Thanks a lot
Eneko
El 12/2/20 a las 16:11, Stefan Reiter escribió:
Based on the RFC and following on- and off-list discussion about
a las 15:58, Stefan Reiter escribió:
Hi,
On 2/13/20 1:27 PM, Eneko Lacunza wrote:
Hi Stefan,
Do you think this could help specifying a custom CPU model that will
migrate VMs with more than 1 core between Intel and AMD CPUs?
https://bugzilla.proxmox.com/show_bug.cgi?id=1660
as discussed
not work, since the requested count might not be
available anymore by the time we want to use them (also, we would then
no longer allocate them correctly on the NUMA nodes).
Add a 'keephugepages' parameter to skip cleanup and simply leave them
untouched.
Signed-off-by: Stefan Reiter
---
I tried adding
y live-migration directionality.
Signed-off-by: Stefan Reiter
---
New in v6.
PVE/API2/Qemu.pm| 9 -
PVE/QemuMigrate.pm | 22 ++
PVE/QemuServer.pm | 14 ++
PVE/QemuServer/CPUConfig.pm | 2 ++
4 files changed, 42 insertions(+
e required).
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 2 +-
PVE/QemuServer/CPUConfig.pm | 73 ++---
2 files changed, 69 insertions(+), 6 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 23176dd..f0c8abd 100644
--- a/PVE
Can be specified for a particular VM or via a custom CPU model (VM takes
precedence).
QEMU's default limit only allows up to 1TB of RAM per VM. Increasing the
physical address bits available to a VM can fix this.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 24
Requires a mock CPU-model config, which is given as a raw string to also
test parsing capabilities. Also tests defaulting behaviour.
Signed-off-by: Stefan Reiter
---
test/cfg2cmd/custom-cpu-model-defaults.conf | 8 ++
.../custom-cpu-model-defaults.conf.cmd| 24
t;runningmachine" and use as override during rollback/resume.
No functional change with non-custom CPU types intended.
Signed-off-by: Stefan Reiter
---
New in v7.
PVE/API2/Qemu.pm | 1 +
PVE/QemuConfig.pm | 36 +++-
PVE/QemuServer.pm | 28 +++
pipermail/pve-devel/2019-July/038268.html
[1]: e.g.:
cpu-model: custom-cpu-name
host-phys-bits 1
flags +aes;+avx;+avx2
reported-model kvm64
[2] https://pve.proxmox.com/pipermail/pve-devel/2020-January/041278.html
qemu-server: Stefan Reiter (8):
Adapt CPUConfig to handle custom models
the test cases to not break. Only the order is
changed, not which flags are present.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 191 +-
test/cfg2cmd/i440fx-win10-hostpci.conf.cmd| 2 +-
test/cfg2cmd/minimal-defaults.conf.cmd| 2
providing a reason why specific CPU flags
have been added, and thus allows for useful warning messages should a
flag be overwritten by another.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 64 +
1 file changed, 64 insertions(+)
diff --git a/PVE/
id writing redundant information to the
config file, additionally get_custom_model is used to retrieve a custom
model configuration by name.
Resolve custom names in print_cpu_device when a custom cpu is passed.
Signed-off-by: Stefan Reiter
---
Depends on pve-cluster including the cpu-models.conf
s to 'input-data', with a size limitation of 1 MiB to
not overwhelm QMP.
Without 'input-data' (API) or '--pass-stdin' (CLI) behaviour is unchanged.
Signed-off-by: Stefan Reiter
---
Tested with:
Debian-VM:
qm guest exec 101 --pass-stdin -- "file" "-" < PVE/QemuServer.pm
On 2/27/20 11:48 AM, Fabian Grünbichler wrote:
On February 26, 2020 11:53 am, Aaron Lauterer wrote:
If IO-Thread is activated and a new enough Qemu version installed the
program still ran into the elsif clause and never in the else clause.
Thus the "include disk ..." log output was missing for
see QEMU commit 744a928cce ("spapr: Stop providing RTAS blob")
Signed-off-by: Stefan Reiter
---
debian/rules | 1 -
1 file changed, 1 deletion(-)
diff --git a/debian/rules b/debian/rules
index dad78dd..2cb0773 100755
--- a/debian/rules
+++ b/debian/rules
@@ -120,7 +120,6 @@ inst
Signed-off-by: Stefan Reiter
---
qemu | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/qemu b/qemu
index 99c5874..b0ca999 16
--- a/qemu
+++ b/qemu
@@ -1 +1 @@
-Subproject commit 99c5874a9b6c9f70aef285d6eff85d4f46de3c52
+Subproject commit
Signed-off-by: Stefan Reiter
---
src/PVE/LXC/Migrate.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/PVE/LXC/Migrate.pm b/src/PVE/LXC/Migrate.pm
index 7df3c4c..6c1e9c1 100644
--- a/src/PVE/LXC/Migrate.pm
+++ b/src/PVE/LXC/Migrate.pm
@@ -284,7 +284,7 @@ sub phase1
on the backend.
Both issues are addressed in this series.
manager: Stefan Reiter (1):
fix #2611: gui: use backendUnit and pretty renderer
www/manager6/dc/OptionView.js | 38 +--
1 file changed, 32 insertions(+), 6 deletions(-)
qemu-server: Stefan Reiter (1):
fix #2611
Signed-off-by: Stefan Reiter
---
PVE/QemuMigrate.pm | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm
index 6db5e62..1753a10 100644
--- a/PVE/QemuMigrate.pm
+++ b/PVE/QemuMigrate.pm
@@ -474,7 +474,7 @@ sub sync_disks
add a pretty renderer that shows units and set backendUnit to submit
correctly converted values.
Signed-off-by: Stefan Reiter
---
www/manager6/dc/OptionView.js | 38 +--
1 file changed, 32 insertions(+), 6 deletions(-)
diff --git a/www/manager6/dc/OptionView.js b
On 2/21/20 9:51 AM, Thomas Lamprecht wrote:
Am 2/20/20 um 6:10 PM schrieb Stefan Reiter:
Warn the user when selecting /dev/random or disabling speed limits.
Note that /dev/random won't be blocking after initialization in a future
kernel release (5.6, IIUC): https://lwn.net/Articles/808575
On 2/21/20 10:13 AM, Thomas Lamprecht wrote:
Am 2/20/20 um 6:10 PM schrieb Stefan Reiter:
Allow a user to add a virtio-rng-pci (an emulated hardware random
number generator) to a VM with the rng0 setting. The setting is
version_guard()-ed.
Limit the selection of entropy source to one of three
For the series:
Tested-by: Stefan Reiter
Reviewed-by: Stefan Reiter
One small thing I noticed is that when Agent is enabled, guest_fstrim
enabled, VM running and one disables the Agent but does not touch the
guest_fstrim box, 'agent: 0' is set to pending, but guest_fstrim is also
On 1/27/20 9:54 AM, Thomas Lamprecht wrote:
On 1/16/20 4:40 PM, Stefan Reiter wrote:
Just like with live-migration, custom CPU models might change after a
snapshot has been taken (or a VM suspended), which would lead to a
different QEMU invocation on rollback/resume.
Save the "-cpu"
The fixme got it right :)
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 9296634..9e00c8a 100644
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
@@ -4865,7 +4865,7 @@ sub
-by: Stefan Reiter
---
PVE/QemuServer.pm | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index ff2e3ee..0c1bd6d 100644
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
@@ -5131,7 +5131,7 @@ sub vmconfig_update_disk {
} else { # cdrom
-by: Stefan Reiter
---
PVE/QemuServer.pm | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 0c1bd6d..abb0528 100644
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
@@ -4504,7 +4504,7 @@ sub qemu_cpu_hotplug {
my $retry
output with units
while at it.
[0] https://qemu.weilnetz.de/doc/qemu-doc.html#Deprecated-features
Signed-off-by: Stefan Reiter
---
PVE/QemuMigrate.pm | 39 ---
1 file changed, 20 insertions(+), 19 deletions(-)
diff --git a/PVE/QemuMigrate.pm b/PVE
is necessary throughout the series.
[*] potentially 2.8, I couldn't find if the 'id' parameter was introduced before
that - not that it matters much
qemu-server: Stefan Reiter (5):
Remove unused $force parameter
qmp: use 'blockdev-change-medium' instead of generic 'change'
qmp: use 'id' parameter
...and cleanup surrounding code a bit.
'change' is deprecated, and according to the qapi definition in QEMU it
is 'strongly recommended' to avoid using it.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 12 +---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/PVE
bytes, duration 35 sec)
_03-17_05_09.vma.zst : 14014 MB...
_03-17_05_09.vma.zst : 14091 MB...
Looks a bit unsightly IMO.
But functionality wise it works fine, tried with a VM and a container, so
Tested-by: Stefan Reiter
for the series.
} else {
die "internal error - un
ately
marked invalid until you select and unselect them) - which is how I noticed.
With this the validation now works correctly.
Signed-off-by: Stefan Reiter
---
Before my last patch they didn't validate enough, now they validate too much...
I tried my best to test all of our ComboGrids once
for ones with a value to be
submitted, i.e. the CorosyncLinkEdit and CorosyncLinkSelector components
are not part of data submission at all.
Change ClusterEdit.js to use the new component for cluster join and
create.
Signed-off-by: Stefan Reiter
---
Kind of depends on the widget-toolkit patch
-devel/2020-January/041168.html
widget-toolkit: Stefan Reiter (1):
ComboGrid: fix on-load validation for blank values
form/ComboGrid.js | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
manager: Stefan Reiter (5):
gui/cluster: validate cluster name length on create
gui/cluster
S itself).
Signed-off-by: Stefan Reiter
---
form/ComboGrid.js | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/form/ComboGrid.js b/form/ComboGrid.js
index 0c5cf1b..54bc239 100644
--- a/form/ComboGrid.js
+++ b/form/ComboGrid.js
@@ -459,10 +459,10 @@ Ext.define('Proxmox.form.C
To help a user identify if they put in the joinInfo for the correct
cluster.
Signed-off-by: Stefan Reiter
---
www/manager6/dc/ClusterEdit.js | 17 +++--
1 file changed, 15 insertions(+), 2 deletions(-)
diff --git a/www/manager6/dc/ClusterEdit.js b/www/manager6/dc/ClusterEdit.js
interface on the new node.
Signed-off-by: Stefan Reiter
---
www/manager6/dc/Cluster.js | 13 +
www/manager6/dc/ClusterEdit.js | 10 --
www/manager6/dc/CorosyncLinkEdit.js | 18 --
3 files changed, 33 insertions(+), 8 deletions(-)
diff --git a/www
Show only links found on the cluster.
'enableColumnHide' is set to false, since it causes confusing behaviour
for the user, considering we'd overwrite part of their choices every 5
seconds.
Signed-off-by: Stefan Reiter
---
www/manager6/Utils.js | 10 +++
www/manager6/dc/Cluster.js | 128
API limits this to 15, validate in GUI as well for instant user
feedback.
Signed-off-by: Stefan Reiter
---
www/manager6/dc/ClusterEdit.js | 1 +
1 file changed, 1 insertion(+)
diff --git a/www/manager6/dc/ClusterEdit.js b/www/manager6/dc/ClusterEdit.js
index a8c021d2..0af858a5 100644
--- a/www
'die' as in 'dice', not what Perl does
Signed-off-by: Stefan Reiter
---
I think it turned out okay ;)
Thanks to everyone who suggested a die as a symbol for randomness.
www/css/ext6-pve.css | 9 ++-
www/images/Makefile | 3 ++-
www/images/icon-die.svg
=f92f96987d7d262047c7604b169a7fdf11236107;hb=HEAD
[2] https://lwn.net/Articles/261804/
[3] https://lwn.net/Articles/808575/
Signed-off-by: Stefan Reiter
---
Includes a version bump currently set to 4.1+pve2. Would need to be changed if
applied later.
Tested with all three options (luckily enough, my system has a hwrng
/808575/
manager: Stefan Reiter (3):
qemu: fix icon alignment in "Add Hardware" menu
qemu: add virtio-rng option to GUI
qemu: change virtio-rng icon to die
www/css/ext6-pve.css | 14 +++-
www/images/Makefile | 3 +-
www/images/icon-die.svg
It was ever so slightly off.
Signed-off-by: Stefan Reiter
---
Couldn't find another place this occured, so I fixed it specifically here.
Not that noticeable currently, but it really felt out of place with the rather
square 'die' icon not being aligned.
www/css/ext6-pve.css | 5
Signed-off-by: Stefan Reiter
---
qm.adoc | 38 ++
1 file changed, 38 insertions(+)
diff --git a/qm.adoc b/qm.adoc
index 0b699e2..3e6d135 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -791,6 +791,44 @@ device of the host use device passthrough (see
uses one hwrng at a time anyway.
Signed-off-by: Stefan Reiter
---
www/manager6/Makefile | 1 +
www/manager6/Utils.js | 2 +-
www/manager6/qemu/HardwareView.js | 23 ++
www/manager6/qemu/RNGEdit.js | 124 ++
4 files changed, 149
On 2/5/20 3:17 PM, Aaron Lauterer wrote:
improve phrasing, align headlines, rearrange requirement lists
Signed-off-by: Aaron Lauterer
---
v2 -> v3:
rebased on current master
v1 -> v2:
applied (some) suggestions from oguz [0], IO performance is based on the
whole system, not just disks
[0]
On 2/5/20 3:17 PM, Aaron Lauterer wrote:
improve phrasing, align style of CLI commands
Signed-off-by: Aaron Lauterer
---
v2 -> v3:
rebased on current master
pve-package-repos.adoc | 143 +
1 file changed, 74 insertions(+), 69 deletions(-)
diff --git
Sorry for being a bit late, seeing as 4 patches are already applied, but
aside from some minor things noted in my other mails: LGTM regarding
style and content!
Reviewed-by: Stefan Reiter
On 2/5/20 3:16 PM, Aaron Lauterer wrote:
This is the first patch series aimed to overhaul our
-by: Stefan Reiter
---
Previous minimum was 0.15, but I have a feeling that even if you managed to
install QEMU 0.15 somehow, it wouldn't have worked anyway.
3.0 is a reasonable requirement IMO, although technically 2.12 would probably
work too, if there's a use-case for using such an outdated
-by: Stefan Reiter
---
v1 -> v2:
* handle error in kvm_user_version
* corrected commit message (2.12 -> 2.8)
@Thomas: If kvm_user_version() failed, we actually died in min_version before.
You're suggested code also didn't quite work, because it still printed warnings
in case $kvmver was
Machines running with SeaBIOS don't have the efidisk attached, so QEMU
cannot back it up and fails with "unknown drive".
Signed-off-by: Stefan Reiter
---
PVE/VZDump/QemuServer.pm | 3 +++
1 file changed, 3 insertions(+)
diff --git a/PVE/VZDump/QemuServer.pm b/PVE/VZDump/QemuServe
Commit 6033f465b8 (gui: pci passthrough: consider domain in PCISelector)
used 'values.host' without checking if it's set, resulting in an error
when the dialog was opened for a new device (where values.host is
undefined).
Signed-off-by: Stefan Reiter
---
www/manager6/qemu/PCIEdit.js | 15
to fix the issue.
Tested-by: Stefan Reiter
Reviewed-by: Stefan Reiter
___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
. EXPECT_ERROR matching is changed to use 'eq' instead of regex
to allow special characters in error messages.
Signed-off-by: Stefan Reiter
---
For some reason, thinking this through is astonishingly confusing. My head is
spinning with QEMU versions, and I'll probably have nightmares of '+pve0'
tonight, but I
Live-migrating a VM with more than 14 SCSI disks to a node that doesn't
support it yet is broken. Use a bumped pve-version to represent that and
give the user a nice error message instead.
Signed-off-by: Stefan Reiter
---
This is more for correctness, since older versions (i.e. ones that don't
with
and without new fallback behaviour.
Signed-off-by: Stefan Reiter
---
I do think that both AS_CLUSTER and AS_JOINEE ages are necessary indeed:
* AS_CLUSTER: If we introduce a parameter to addnode that we cannot have a
fallback for, and old versions don't provide it, we can exclude them from
lock code that strips extra information from an Exception
instance. This ensures that multi-line formatted errors can be returned.
Warnings are always returned as array, to be printed on the caller.
Includes testing.
Signed-off-by: Stefan Reiter
---
data/PVE/API2/ClusterConfig.p
ven as constant MAX_LINK_COUNT, should it
change in the future.
All necessary functions have been updated to
use the new $links array format instead of seperate $link0/$link1
parameters, and call sites changed accordingly.
Signed-off-by: Stefan Reiter
---
This patch intentionally removes some verificat
probably more discussion-worthy ;)
* changed join-API concept to include two API ages (see patch 5 for more) - as
discussed off-list with Thomas
* rebase on master
cluster: Stefan Reiter (4):
corosync: add verify_conf
Enable support for up to 8 corosync links
Add verification and fallback
cluster link.
Signed-off-by: Stefan Reiter
---
data/PVE/API2/ClusterConfig.pm | 63 +-
data/PVE/CLI/pvecm.pm | 7
data/PVE/Cluster/Setup.pm | 7
3 files changed, 76 insertions(+), 1 deletion(-)
diff --git a/data/PVE/API2/ClusterConfig.pm b
-by: Stefan Reiter
---
Not sure if we usually include fixes like that, but I feel like this could avoid
a lot of Forum threads once TR 3000 gets more commonplace :)
...w-Reserved-types-to-be-overwritten-i.patch | 88 +++
1 file changed, 88 insertions(+)
create mode 100644
patches
get_basic_machine_info was removed by commit 045749f2fc.
Use get_host_arch to get the default machine type instead, and
optionally allow to specify architecture as parameter.
Signed-off-by: Stefan Reiter
---
@Thomas: The mentioned commit happened after I sent v6, so the patch you merged
would
providing a reason why specific CPU flags
have been added, and thus allows for useful warning messages should a
flag be overwritten by another.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 64 +
1 file changed, 64 insertions(+)
diff --git a/PVE/
id writing redundant information to the
config file, additionally get_custom_model is used to retrieve a custom
model configuration by name.
Signed-off-by: Stefan Reiter
---
Depends on updated pve-cluster.
PVE/QemuServer/CPUConfig.pm | 109 +++-
1 file ch
Can be specified for a particular VM or via a custom CPU model (VM takes
precedence).
QEMU's default limit only allows up to 1TB of RAM per VM. Increasing the
physical address bits available to a VM can fix this.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 24
e required).
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 2 +-
PVE/QemuServer/CPUConfig.pm | 73 ++---
2 files changed, 69 insertions(+), 6 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 63fe25c..ee95b31 100644
--- a/PVE
.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm | 219 +-
PVE/QemuServer/CPUConfig.pm | 232
PVE/QemuServer/Makefile | 1 +
3 files changed, 235 insertions(+), 217 deletions(-)
create mode 100644 PVE/QemuServer
Requires a mock CPU-model config, which is given as a raw string to also
test parsing capabilities. Also tests defaulting behaviour.
Signed-off-by: Stefan Reiter
---
test/cfg2cmd/custom-cpu-model-defaults.conf | 8 ++
.../custom-cpu-model-defaults.conf.cmd| 24
ucture
* Add and fix test cases
Does not include any fix for versioning/live-migration with custom models for
now, felt these changes were big enough. Also applies to CPU hotplug with custom
models.
[0]: https://pve.proxmox.com/pipermail/pve-devel/2019-July/038268.html
[1]: e.g.:
cpu-model: custom-c
t;runningmachine" and use as override during rollback/resume.
No functional change with non-custom CPU types intended.
Signed-off-by: Stefan Reiter
---
New in v7.
PVE/API2/Qemu.pm | 1 +
PVE/QemuConfig.pm | 36 +++-
PVE/QemuServer.pm | 28 +++
the test cases to not break. Only the order is
changed, not which flags are present.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 191 +-
test/cfg2cmd/i440fx-win10-hostpci.conf.cmd| 2 +-
test/cfg2cmd/minimal-defaults.conf.cmd| 2
y live-migration directionality.
Signed-off-by: Stefan Reiter
---
New in v6.
PVE/API2/Qemu.pm| 9 -
PVE/QemuMigrate.pm | 22 ++
PVE/QemuServer.pm | 14 ++
PVE/QemuServer/CPUConfig.pm | 2 ++
4 files changed, 42 insertions(+
Holds information about custom CPU models, section config style. Used in
qemu-server.
Signed-off-by: Stefan Reiter
---
data/PVE/Cluster.pm | 2 +-
data/src/status.c | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/data/PVE/Cluster.pm b/data/PVE/Cluster.pm
index b4fac03
On 1/10/20 4:12 PM, Thomas Lamprecht wrote:
Allow to collapse and expand the "Notes" panel on the virtual guests
summary panel.
Further add a browser setting to control the default behavior, one
can chose from:
* expand on show (default)
* collapse on show
* collapse if there are no notes
Signed-off-by: Stefan Reiter
---
Thanks for reminding me :)
v2 includes Aarons review as well as feedback I discussed with him off-list
qm.adoc | 38 ++
1 file changed, 38 insertions(+)
diff --git a/qm.adoc b/qm.adoc
index 0b699e2..cd6641f 100644
Upstream marks these as having a micro-version of >=90, unfortunately the
machine versions are bumped earlier so testing them is made unnecessarily
difficult, since the version checking code would abort on migrations etc...
Signed-off-by: Stefan Reiter
---
PVE/QemuServer.pm |
ave it set but an incompatible controller), but do warn
them that the setting will be ignored.
Signed-off-by: Stefan Reiter
---
www/manager6/qemu/HDEdit.js | 28 +++-
1 file changed, 27 insertions(+), 1 deletion(-)
diff --git a/www/manager6/qemu/HDEdit.js b/www/man
via JSONSchema directly.
With those changes it became visible that the 'phys-bits' format was
actually invalid, so add a format_description to fix it.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 60 -
1 file changed, 19 insertions(+), 41 deleti
get_format consistently to avoid future breakages.
No existing functionality is intentionally changed.
Signed-off-by: Stefan Reiter
---
@Fabian G.: Since we discussed this off-list, is this good? At least there
shouldn't be a possibility for a parser/hash format drift the way I've
implemented
the test cases to not break. Only the order is
changed, not which flags are present.
Signed-off-by: Stefan Reiter
Reviewed-By: Fabian Ebner
Tested-By: Fabian Ebner
---
PVE/QemuServer/CPUConfig.pm | 191 +-
test/cfg2cmd/efi-raw-old.conf.cmd | 2
Can be specified for a particular VM or via a custom CPU model (VM takes
precedence).
QEMU's default limit only allows up to 1TB of RAM per VM. Increasing the
physical address bits available to a VM can fix this.
Signed-off-by: Stefan Reiter
---
Changes in v10:
* Change phys-bits format
providing a reason why specific CPU flags
have been added, and thus allows for useful warning messages should a
flag be overwritten by another.
Signed-off-by: Stefan Reiter
Reviewed-By: Fabian Ebner
Tested-By: Fabian Ebner
---
PVE/QemuServer/CPUConfig.pm | 64 +++
Requires a mock CPU-model config, which is given as a raw string to also
test parsing capabilities. Also tests defaulting behaviour.
Signed-off-by: Stefan Reiter
Reviewed-By: Fabian Ebner
Tested-By: Fabian Ebner
---
Changes in v10:
* Added test for new phys-bits format
(Since it's just
y live-migration directionality.
Signed-off-by: Stefan Reiter
---
Changes in v10:
* Introduce and use PVE::QemuServer::CPUConfig::get_cpu_from_running_vm helper
PVE/API2/Qemu.pm| 7 +++
PVE/QemuMigrate.pm | 15 +++
PVE/QemuServer.pm | 13 +++
t;runningmachine" and use as override during rollback/resume.
No functional change with non-custom CPU types intended.
Signed-off-by: Stefan Reiter
---
Changes in v10:
* Use PVE::QemuServer::CPUConfig::get_cpu_from_running_vm helper
PVE/API2/Qemu.pm | 1 +
PVE/QemuConfig.pm | 26 +++
flags +aes;+avx;+avx2
reported-model kvm64
[2] https://pve.proxmox.com/pipermail/pve-devel/2020-January/041278.html
qemu-server: Stefan Reiter (6):
Include "-cpu" parameter with live-migration
Include "-cpu" parameter with snapshots/suspend
Add helpers to
ping, I think we'd want this in 6.2?
On 23/03/2020 13:41, Stefan Reiter wrote:
v2 -> v3:
* add patch 1 (localization fix)
* implement changes from Dominik's review:
* use 'let' in new code
* use references for element lookup
* some code style nits
* fix formatting (simp
On 21/04/2020 16:01, Dominik Csapak wrote:
netdev_add is now a proper qmp command, which means that it verifies
the parameter types properly
The patch looks reasonable, but I'm confused how that worked before?
'git blame' tells me the typed QMP schema for netdev_add has been in
QEMU since
On 21/04/2020 16:43, Dominik Csapak wrote:
On 4/21/20 4:38 PM, Stefan Reiter wrote:
On 21/04/2020 16:01, Dominik Csapak wrote:
netdev_add is now a proper qmp command, which means that it verifies
the parameter types properly
The patch looks reasonable, but I'm confused how that worked
On 07/04/2020 09:04, Fabian Ebner wrote:
On 06.04.20 15:01, Stefan Reiter wrote:
On 06/04/2020 14:31, Fabian Ebner wrote:
On 26.03.20 16:13, Stefan Reiter wrote:
Just like with live-migration, custom CPU models might change after a
snapshot has been taken (or a VM suspended), which would lead
ays
initialized from the defaults anyway.
Signed-off-by: Stefan Reiter
---
I am not sure when this code is used at all... I tried slowing the network down
to 56k-modem levels, and migration still completed without triggering it (both
with default migrate_downtime and migrate_downtime set to 1ms).
PVE/Q
; see [2] for older history >
[0]: https://pve.proxmox.com/pipermail/pve-devel/2019-July/038268.html
[1]: e.g.:
cpu-model: custom-cpu-name
host-phys-bits 1
flags +aes;+avx;+avx2
reported-model kvm64
[2] https://pve.proxmox.com/pipermail/pve-devel/2020-January/041278.html
qemu-se
y live-migration directionality.
Signed-off-by: Stefan Reiter
---
PVE/API2/Qemu.pm| 8 +++-
PVE/QemuMigrate.pm | 21 +
PVE/QemuServer.pm | 13 +
PVE/QemuServer/CPUConfig.pm | 2 ++
4 files changed, 39 insertions(+), 5 deletions
Requires a mock CPU-model config, which is given as a raw string to also
test parsing capabilities. Also tests defaulting behaviour.
Signed-off-by: Stefan Reiter
---
test/cfg2cmd/custom-cpu-model-defaults.conf | 8 ++
.../custom-cpu-model-defaults.conf.cmd| 24
t;runningmachine" and use as override during rollback/resume.
No functional change with non-custom CPU types intended.
Signed-off-by: Stefan Reiter
---
PVE/API2/Qemu.pm | 1 +
PVE/QemuConfig.pm | 34 ++
PVE/QemuServer.pm | 28
3 files
providing a reason why specific CPU flags
have been added, and thus allows for useful warning messages should a
flag be overwritten by another.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 64 +
1 file changed, 64 insertions(+)
diff --git a/PVE/
Can be specified for a particular VM or via a custom CPU model (VM takes
precedence).
QEMU's default limit only allows up to 1TB of RAM per VM. Increasing the
physical address bits available to a VM can fix this.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 24
the test cases to not break. Only the order is
changed, not which flags are present.
Signed-off-by: Stefan Reiter
---
PVE/QemuServer/CPUConfig.pm | 191 +-
test/cfg2cmd/i440fx-win10-hostpci.conf.cmd| 2 +-
test/cfg2cmd/minimal-defaults.conf.cmd| 2
301 - 400 of 476 matches
Mail list logo
