ma': 'str'} }
>> +
>> +##
>> +# @query-config-schema
>> +#
>> +# Query configuration schema information of options
>
> And this.
But here, it might make sense to call out that the default is all
options, but specifying a name limits to one option.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ady decided that query-* commands can add
output fields to their dictionary as needed).
The name 'config-schema' sounds like it is machine-parseable; can we use
a more descriptive name such as 'help-text' that makes it obvious what
you are really providing?
> +++ b/qmp.c
> @@ -24,6 +24,9 @@
> #include "hw/qdev.h"
> #include "sysemu/blockdev.h"
> #include "qom/qom-qobject.h"
> +#include "qemu-options.h"
> +#include "net/net.h"
What does net/net.h have to do with anything?
Overall, looks quite useful, and it is worth getting into 1.5.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
JSON
representation of what the command expects, THAT would make it
acceptable to add optional arguments to an existing command in a way
that libvirt could query to see whether the qemu it is talking to
supports the optional arguments.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvir
nclined to NACK this unless you can give better explanation why
send-key should behave differently than bare metal. If anything, the
behavior being complained about is a "feature" of the console of the
guest being tested, not something where we should change how the
hardware be
to
-netdev, we should also convert -netdev to qemuOpts at that time, so
that libvirt would know whether the new feature is available.
At any rate, we really DO want introspection, and having it in 1.5 is a
worthwhile goal. Even if the introspection turns up empty on legacy
options, having it fo
me = file->filename;
> +} else {
> +filename = qdict_get_try_str(options, "filename");
> +}
> +
> +trace_bdrv_open_common(bs, filename ?: "", flags, drv->format_name);
Yet another case of a non-C99 gcc extension; but as this is not the
first use, I&#
On 04/22/2013 01:32 AM, Amos Kong wrote:
> On Sat, Apr 20, 2013 at 10:06:28AM -0600, Eric Blake wrote:
>> On 04/18/2013 10:44 PM, Amos Kong wrote:
>>> (qemu) sendkey a 1000
>>>
>>> Current design is that qemu only send one down event to guest,
>>> a
lp if we had introspection of QMP commands?
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
with fdsets, then libvirt
can already achieve pass to pipe - and pass to pipe is indeed the nicest
way to then get alternate formats without having to decode lots of JSON,
since there are conversion utilities that can take ppm on input and
output other formats without having to teach qemu those other
in active use, then disabling this feature will cause all 8GB to
Trailing whitespace here and elsewhere. scripts/checkpatch.pl doesn't
like that.
s/disabling/enabling/
> +be pinned and resident in memory. This feature mostly affects the
> +bulk-phase round of the migration and can be
f->ops->get_max_size(f, f->opaque,
> +transferred_bytes, time_spent, max_downtime);
> +}
> +
> +return ((double) (transferred_bytes / time_spent)) *
> +max_downtime / 100;
Cast to double is too late; you've already suffered from integer
divisio
> footprint is
Trailing whitespace, and line longer than 80 columns. You ought to
rewrap this, and make sure it passes checkpatch.pl.
But since that is whitespace-only, feel free to add:
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ax: 1
> item[7].features: 126614521
And this would then be returned as a JSON array containing struct
members looking like this:
> +{ 'type': 'X86CPUFeatureWordInfo',
> + 'data': { 'cpuid-input-eax': 'int',
> +'*cpuid-input-ecx
e need to
expand the 'MachineInfo' QMP datatype so that 'query-machines' can tell
us whether a given machine will support or reject attempts to set
'paging':true during 'dump-guest-memory'?
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 04/23/2013 07:20 AM, Luiz Capitulino wrote:
> On Mon, 22 Apr 2013 09:00:05 -0600
> Eric Blake wrote:
>
>> At any rate, we really DO want introspection, and having it in 1.5 is a
>> worthwhile goal. Even if the introspection turns up empty on legacy
>> options, ha
nd than to add parameters to the existing
one. At any rate, we're late enough that 'screendump' probably won't be
improved in 1.5, so we have the full 1.6 cycle to get it right.
Not on the phone call, but a recent mail that is related to the topic -
feature detection of whethe
should be used.
>
> There was a proposal for enumerating possible CPUs in previous version,
> but it was target specific, and I was convinced to drop it for 1.5 and
> aim for generic way to expose this information later.
> http://lists.gnu.org/archive/html/qemu-devel/2013-04/msg02286.html
&g
we don't repeat that mistake, we need to
decide whether this should still go into 1.5 with a plan of adding
parameters for 1.6, or whether we should add a counterpart query-*
command that makes it easy to determine how much of block-backup is
supported, or [your suggestion here]
--
Er
The existing drive-mirror command can also target an NBD destination,
with similar effects.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
e series
that interacts with my current interests (I tend to review anything QMP,
because I want to make sure the design will be sane for libvirt
interaction, while overlooking things like migration internals because
they are black box ops to libvirt if the interface was sane).
--
Eric Blake
osed to be such a
list; and since it's a wiki, feel free to add things that you learned
here that weren't adequately covered yet.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 04/23/2013 03:44 PM, Anthony Liguori wrote:
> Eric Blake writes:
>
>> On 04/23/2013 02:15 PM, Michael R. Hines wrote:
>>>
>>> Would be helpful for new people (like myself) to have a summary
>>> of these procedures on the wiki so we don't have to b
t it definitely makes more sense if the entire
command is waiting for 1.6, since we've already entered soft freeze for 1.5.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
---
> blockdev.c | 118
> +--
> tests/qemu-iotests/group | 2 +-
> 2 files changed, 115 insertions(+), 5 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.
> ]
>
> Signed-off-by: Amos Kong
> ---
> I used an _ugly_ "#ifndef" to limit it only be defined in qapi-types.h.
> do you have some suggestion to fix it?
I'm not the maintainer, so you may still want to wait for a better
suggestion. But since I personally
option, Error
> **errp)
> +{
> +ConfigSchemaInfoList *conf_list = NULL, *entry;
> +ConfigSchemaInfo *info;
> +strList *str_list = NULL, *str_entry;
> + int entries, i, j;
> +
> +entries = ARRAY_SIZE(vm_config_groups);
> +
> +for (i = 0; i < entries; i++) {
>
On 04/24/2013 06:47 AM, Amos Kong wrote:
> Currently we can only use ['String'] to add string to a list,
> it contains some additional JSON structure.
Side note - when sending a 2-patch series, it helps to include a 0/2
cover letter.
--
Eric Blake eblake redhat com+1-919-
ary, rather than opening
n emails just to see if it touches any file I care about). I guess that
means I'm asking for a bit of help in further updating the wiki to
mention how to have git reliably include a cover letter diffstat, since
asking for something without giving a recipe of how to ac
ists of 3 bits, each bit represents the Caps, Num, and Scroll lock
> +key respectively. '1' indicates that the LED should be on and '0' should
> +be off.
Which bit is which? Is bit 0 the Caps lock LED?
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
flicts regarding the algorithm for snapshot
lookup, and whether there are any rebase dependencies between your
series and Wenchao's.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
t is easy to backport, even if the introspection does not go in
at the same time as the original feature, where improved error messages
becomes a quality-of-implementation value-added by distro packagers.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
I think returning EXIT_SUCCESS/EXIT_FAILURE
instead of 0/1 is a bit nicer at expressing why we chose a positive
value; but that would be a separate cleanup to all of qemu-img.c.
Hence, I have no problems giving:
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt vir
ave
hindered the ability to learn which options exist even when you didn't
know their name a priori.
>
>> + 'returns': ['ConfigSchemaInfo']}
>> diff --git a/qmp-commands.hx b/qmp-commands.hx
>> index 4d65422..19415e4 100644
>> --- a/qmp-
gt; +# @cpu-add
> +#
> +# Adds CPU with specified ID
> +#
> +# @id: ID of CPU to be created, valid values [0..max_cpus)
> +#
> +# Returns: Nothing on success
Missing a 'Since 1.5' designation.
> +##
> +{ 'command': 'cpu-add', 'data': {'id&
for exposing the paging boolean. At the internal C level,
qemuMonitorJSONDump hardcodes 'paging' to false in current libvirt.git.
I did a bit more digging, and found this libvirt commit:
commit d239085e956ca6ca42480e877e98a4302e91b853
Author: Eric Blake
Date: Mon Sep 17 13:05:29 201
ar *snapshot_id,
> + Error **errp)
> {
> /* FIXME: Delete specified snapshot id. */
> -return 0;
> +error_setg(errp, "Deleting snapshot is not supported");
Commit message should mention this bug fix.
But the code looked okay to me, so:
HAVE to rush this into 1.5;
libvirt doesn't (yet) expose the buf-size or granularity options on to
the end user (although I do have plans to get to that point); and Paolo
pointed out that the try-and-fail approach (omit the argument if the
user requests the default, and try the argument relying on qemu
t; g_free(sn_tab);
> -return ret;
> +return found;
return *sn_info != NULL;
> }
If you _do_ decide to keep the boolean variable instead of hard-coding a
false return and avoiding redundancy by using other variables to
determine the result, then at least s/found/ret/, because I
On 04/23/2013 10:06 AM, Eric Blake wrote:
>>
>> we can change "drive_mirror" to use a new command to see if there
>> are the new features.
>
> drive-mirror changed in 1.4 to add optional buf-size parameter; right
> now, libvirt is forced to limit it
that
looks something like { 'name':'str', 'type':'',
'optional':'boolean', '*default':'', '*documentation':'str' }. If we don't
provide filtering, then this JSON output might be quite large, because
i
On 04/24/2013 03:59 PM, Eric Blake wrote:
> With full-blown command introspection, we would want to be asking for a
> JSON representation more details about each parameter, in a struct that
> looks something like { 'name':'str', 'type':'',
&g
+},
> +
> +SQMP
> +vm-snapshot-delete
> +--
> +
> +Delete a snapshot identified by name or id or both. One of the name or id
> +is required. It will returns SnapshotInfo of successfully deleted snapshot.
s/returns/return/
> @@ -2556,31 +2557,73 @@ int load_vmstate(const char *name)
> return 0;
> }
>
> -void do_delvm(Monitor *mon, const QDict *qdict)
> +SnapshotInfo *qmp_vm_snapshot_delete(const bool has_name, const char *name,
> + const bool has_id, const char *id,
> + Error **errp)
Actual function looks right to me.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 04/24/2013 01:39 PM, Eric Blake wrote:
> On 04/24/2013 12:20 PM, Luiz Capitulino wrote:
>> On Thu, 25 Apr 2013 01:33:24 +0800
>> Amos Kong wrote:
>>
>>> Libvirt has no way to probe if an option or property is supported,
>>> This patch introdues a new qm
ring for later,
but I'd rather not have mandatory 'option' as then you have to know in
advance which command line options you care about.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
y been
switching to default-free coding styles on any switch statement where we
expect an enum will grow in the future, for that reason.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
?
Add a new capability to the QMP enum MigrationCapability; default it to
off, and let people use the existing QMP command
migrate-set-capabilities to request the new auto-converge mode.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
s
On 04/24/2013 08:25 PM, Lei Li wrote:
> On 04/24/2013 10:57 PM, Eric Blake wrote:
>> On 04/24/2013 04:12 AM, Lei Li wrote:
>>> Signed-off-by: Lei Li
>>> ---
>>> docs/vnc-ledstate-Pseudo-encoding.txt | 40
>>> +++
gt;>> +return 1;
>>> +}
>>> +return 0;
>>
>> Maybe it's just me, but I think returning EXIT_SUCCESS/EXIT_FAILURE
>> instead of 0/1 is a bit nicer at expressing why we chose a positive
>> value; but that would be a separate cleanup to
}
But I don't know what the existing callers need, to make a decision on
whether a signature change is warranted. Again, more reason to defer
this series to 1.6.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ookup
cleaner, removing hard-coded length limits on HMP snapshot display.
Yes, getting the common parts clean as one series, then doing two more
relatively-independent series of 1. better query output, 2. QMP
counterpart to snapshot manipulations, is probably workable.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ime to fix this now.
Agreed - we've added additional information in the QMP types returned by
query-* in the past, and it's no problem if we don't expose default
values via QMP until improving the C code to track default values in
1.6. For 1.5, let's just focus on exposing
hat is in-use by a qemu process, and instead only
use qemu monitor commands; since qemu-img may see inconsistent state,
and worse, any writes that qemu-img do may cause qemu to see
inconsistent state and cause corruption to the running guest.
--
Eric Blake eblake redhat com+1-919-301-326
ming Luiz can fix them up while
putting on the QMP pull request, without needing to see a v6.
Reviewed-by: Eric Blake
> +##
> +# @CommandLineParameterType:
> +#
> +# Possible types for an option parameter.
> +#
> +# @string: accepts a character string
> +#
> +# @boolean: accepts
On 04/19/2013 11:47 AM, Luiz Capitulino wrote:
> Next commit wants to use it.
>
> Signed-off-by: Luiz Capitulino
> ---
> block.c | 3 +--
> include/block/block.h | 1 +
> 2 files changed, 2 insertions(+), 2 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blak
ve_del(). This fixes the problem
> commit 9ca1115 intended to fix, plus the spurious events.
>
> Signed-off-by: Luiz Capitulino
> ---
> block.c| 2 --
> blockdev.c | 2 ++
> 2 files changed, 2 insertions(+), 2 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake
t;invalid option name: %s", option);
>
> But I can fix that myself in qmp branch.
Based on include/qapi/qmp/qerror.h, the message you would be replacing is:
#define QERR_INVALID_OPTION_GROUP \
ERROR_CLASS_GENERIC_ERROR, "There is no option group '%s'"
Yeah, "invalid option name: foo" reads better than "There is no option
group 'foo'".
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
unction where its content isn't needed any more.
>
> Signed-off-by: Kevin Wolf
> ---
> block/rbd.c | 9 ++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization librar
ream to worry
about, and shouldn't affect what we put into 1.5 upstream).
And I just realized: this patch is queued to go into 1.5, and if I read
it correctly, then once applied, -drive will no longer be empty in
query-command-line-options output:
https://lists.gnu.org/archive/html/qemu-devel/
| 10 +++---
> block/sheepdog.c | 18 --
> include/block/block.h | 5 +++--
> include/block/block_int.h | 5 +++--
> qemu-img.c| 7 ++-
> savevm.c | 10 +-
> 9 files changed, 66 insertions(+), 51 d
POSIX allows but not requires this:
http://pubs.opengroup.org/onlinepubs/9699919799/functions/V2_chap02.html#tag_15_02_02
says that the str* prefix is reserved in ), a
strictly-conforming implementation cannot rely on this extension, and
there are systems where you have to get strcasecmp() from ,
just
-
> include/block/block.h | 3 ++-
> include/block/block_int.h | 3 ++-
> qemu-img.c| 11 ---
> savevm.c | 15 +--
> 9 files changed, 60 insertions(+), 32 deletions(-)
>
Reviewed-by: Eric Blake
--
Eric Blake eblake
to 1.6 when
we can get the entire design right, than to have 1.5 expose something
that can't be reliably used and which might change based on designing
the rest of the improvements.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
his code is also dumb.
>
> Ok, but since the discussion is in public list, no need for CVE then.
Wrong. CVEs are useful even for publicly disclosed bugs. It tells
people whether they need to upgrade in order to avoid a vulnerability.
What we don't need is embargo. But we do need a CVE
cent query-command-line-options, putting the reminder
comment to keep QMP and C in sync only in the C file.)
> +#
> +# Since: 1.5
This missed 1.5; it will need to be 1.6 now.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
uch a bad file can only be created
externally). The goal of a commit message is to convince the readers
that a change in semantics is appropriate because the alternative of
leaving things broken is worse.
[1]https://lists.gnu.org/archive/html/qemu-devel/2013-04/msg03501.html
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
iles changed, 142 insertions(+), 132 deletions(-)
>
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
image file format
> +Format of image file. It must be non-empty. It
> +is coded in free-form ASCII, and is not
> +NUL-terminated. Zero padded on the right.
Again, requiring a format contradicts the earlier statement about format
probing.
How does this compare with Paolo's efforts to design a persistent bitmap
for drive-mirror/block-backup use?
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
Returns: nothing on success
> +# If @device is not a valid block device, DeviceNotFound
> +#
> +# Since 1.5
1.6, now
> +##
> +{ 'command': 'block-backup',
> + 'data': { 'device': 'str', 'target': 's
On 04/26/2013 04:52 PM, Eric Blake wrote:
> On 04/23/2013 10:25 AM, Stefan Hajnoczi wrote:
>> @block-backup
[hit send too soon]
Subject line should mention block-backup, not block_backup.
>> +##
>> +{ 'command': 'block-backup',
>> + 'data&#
b
related events in QMP/qmp-events.txt should be updated to refer to the
enum instead of also being open-coded 'str'. Will this job be called
"backup"?
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
monitor are both processed from the
> +same event loop which monitors their file descriptors until they become
s/loop/loop,/
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
++
> 2 files changed, 65 insertions(+), 20 deletions(-)
>
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
to QAPI actually let you do
> that with a QAPI union while keeping backwards-compatible output for
> other fields. Ok to do this later? It should be acceptable for soft
> freeze.
Correct, use of a discriminated union can add a new 'type' parameter,
which in turn controls w
QEMU_ARCH_ALL)
But this implies both parameters can be used at once. Is it worth
rewriting in two lines:
"-numa node[,mem=size][,cpus=cpu[-cpu]][,nodeid=node]\n"
"-numa node[,memdev=id][,cpus=cpu[-cpu]][,nodeid=node]\n"
to make the exclusion clearer?
> -to allocate RAM and vCPUs respectively.
> +to allocate RAM and vCPU srespectively, and possibly @option{-object}
s/vCPU srespectively/vCPUs respectively/
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
+Show memory devices information.
> +
> +
> +Example (1):
> +
> +-> { "execute": "query-memdev" }
> +<- { "return": [
> + {
> + "size": 536870912,
> + "host-nodes": [0, 1],
> + "policy": "bind"
> + },
...and then only demonstrate 3 in the example. Something's not quite right.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
xi| 2 +-
> 3 files changed, 25 insertions(+), 5 deletions(-)
>
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
gt; qemu-img-cmds.hx | 4 ++--
> qemu-img.c | 24 +---
> qemu-img.texi| 9 -
> 3 files changed, 27 insertions(+), 10 deletions(-)
>
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library ht
may have been shorter to write, but this does
indeed appear to do the trick for all but a perverse person that names
their backing files with substrings that include something like
'/path/to/Mapped to gotcha/'.
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ts/qemu-iotests/group | 1 +
> 3 files changed, 242 insertions(+)
> create mode 100755 tests/qemu-iotests/094
> create mode 100644 tests/qemu-iotests/094.out
>
The commit message can be touched up by whoever does the PULL request, so:
Reviewed-by: Eric Blake
--
Eric Blak
sts/qemu-iotests/095.out | 26 +
> tests/qemu-iotests/group | 1 +
> 3 files changed, 99 insertions(+)
> create mode 100755 tests/qemu-iotests/095
> create mode 100644 tests/qemu-iotests/095.out
>
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat c
@
> ##
> { 'command': 'drive-mirror',
>'data': { 'device': 'str', 'target': 'str', '*format': 'str',
> +'*new-node-name': 'str', '*to-replace-node-name'
char *name'. I am casting to 'char *
> now', is it better?
Okay, I see. Sometimes, I like to put in a comment /* cast away const
*/ to make it obvious why I'm doing things like that. It also serves as
a spur to investigate a couple of things: 1) should the callback
signa
n",
error_report() should not include trailing \n
> + numa_total, ram_size);
> +exit(1);
Not your fault that this file is full of exit(1), but it would be nice
to have a cleanup patch someday that uses EXIT_FAILURE.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 06/09/2014 08:29 PM, Hu Tao wrote:
> On Mon, Jun 09, 2014 at 05:02:51PM -0600, Eric Blake wrote:
>> On 06/09/2014 04:25 AM, Hu Tao wrote:
>>> From: Wanlong Gao
>>>
>>> If the total number of the assigned numa nodes memory is not
>>> equal to the as
:'str',
> +'*vhost-force':'bool' } }
[1]...and here you call it vhost-force...
> +Establish a vhost-user netdev, backed by a chardev @var{id}. The chardev
> should
> +be a unix domain socket backed one. The vhost-user uses a specifical
On 06/10/2014 03:20 AM, arei.gong...@huawei.com wrote:
> From: Gonglei
>
> Signed-off-by: ChenLiang
> Signed-off-by: Gonglei
> ---
> qobject/json-parser.c | 15 ++-
> 1 file changed, 2 insertions(+), 13 deletions(-)
Reviewed-by: Eric Blake
> @@ -461,10
-only commands are allowed. Luiz?
Correct - please expose this through QMP.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
pected flag
combinations as invalid, even when each flag is in isolation okay,
rather than the current behavior of silently prioritizing one flag and
ignoring the other?
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 06/10/2014 06:55 AM, Eric Blake wrote:
> On 06/10/2014 03:29 AM, Peter Lieven wrote:
>> if a saved vm has unknown flags in the memory data qemu
>> currently simply ignores this flag and continues which
>> yields in an unpredictable result.
>>
>> This patch cat
On 06/10/2014 07:03 AM, Michael S. Tsirkin wrote:
> On Tue, Jun 10, 2014 at 06:11:16AM -0600, Eric Blake wrote:
>> On 06/10/2014 04:02 AM, Nikolay Nikolaev wrote:
>>> The supplied chardev id will be inspected for supported options. Only
>>> a socket backend, with a s
ed-off-by: Nikolay Nikolaev
> ---
> net/vhost-user.c | 18 ++
> qapi-schema.json |4 ++--
> 2 files changed, 8 insertions(+), 14 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http:
nt, since we still don't have global qapi introspection, the
addition of a new command is nicer than the addition of an optional
parameter.
But I definitely agree that we have a discrepancy between HMP being able
to specify a current CPU, vs. QMP being unable to direct which CPU gets
the interrupt.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
gt; 2 files changed, 22 insertions(+)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
es
> management software.
>
> Let's do the same thing when seeing a 0-length image: report ready
> immediately; wait for block-job-cancel or block-job-complete; clear the
> cancel flag as existing non-zero image synced case (cancelled after
> ready); then jump to the exit.
041.out | 4 ++--
> 2 files changed, 8 insertions(+), 5 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
> inject-nmi
> --
>
> -Inject an NMI on guest's CPUs.
> +Inject an NMI on the default CPU (x86/s390) or all CPUs (ppc64).
Oh, I see the problem. The acronym 'NMI' is pronounced with a leading
vowel (an "en-em-eye"), but the expansion is pronounced with a consonant
(a non-maskable interrupt), so the correct article depends on whether
you are spelling things out.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
drv = bdrv_find_format(drvname);
...and this does it again. Should this second assignment be inside an
else{}?
> +static int bdrv_file_open(BlockDriverState *bs, const char *filename,
> + QDict **options, int flags, Error **errp)
> +
> +drv = bdrv_find_format(drvname);
> +if (!drv) {
> +error_setg(errp, "Unknown driver '%s'", drvname);
> +ret = -ENOENT;
> +goto fail;
> +}
Isn't this error check redundant with the one already done in
bdrv_fill_options()? You could probably just use assert(drv)
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
Dict **options, const char *filename,
> +static int bdrv_fill_options(QDict **options, const char *filename, int
> flags,
> Error **errp)
Ah, this fixes one of my findings on 1/9.
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-91
h the file name from the options QDict if necessary */
> -if (filename) {
> +if (protocol && filename) {
> if (filename && !qdict_haskey(*options, "filename")) {
Slight conflict here when you clean up the dead 'filename &&' in patch
1;
On 06/11/2014 08:04 AM, Kevin Wolf wrote:
> Signed-off-by: Kevin Wolf
> ---
> block.c | 88
> +
> 1 file changed, 45 insertions(+), 43 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com
901 - 1000 of 23401 matches
Mail list logo
