On 02/11/2013 10:05 AM, Markus Armbruster wrote:
> Signed-off-by: Markus Armbruster
> ---
> qapi-schema.json | 9 ++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization
nable name before 1.4 is cut, I don't see a problem with changing
the name. It's just that time is not on our side.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
st.
> +##
> +{ 'command': 'migrate_set_rdma_port', 'data': {'port': 'int'} }
> +
> +##
> +# @migrate_set_rdma_host
> +#
> +# Set rdma communication host address.
> +#
> +# @host: rdma communication host address.
> +#
> +#
cache-size',
'query-migrate-cache-size'? I argue that 'migrate_cancel' and
'migrate_set_downtime' are the odd ones out, and only because they were
written before our current policy of preferring dash.
> Shouldn't I stay consistent with the other command
table @option
> +
> +@item 0
> +Images are identical
> +@item 1
> +Images differ
> +@item 2
> +Error on opening an image
> +@item 3
> +Error on checking a sector allocation
> +@item 4
> +Error on reading data
> +
> +@end table
--
Eric Blake eblake re
re we can
> do that, too bad, we'll live.
>
> Please consider seriously.
Seems reasonable to me to get this into 1.4.
Series: 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
is one way to easily do this).
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
+++---
> memory.c | 4 +-
> target-i386/topology.h| 6 +--
> util/bitops.c | 6 +--
> util/hbitmap.c| 3 +-
> 7 files changed, 112 insertions(+), 104 deletions(-)
Series:
Reviewed-by: Eric Blake
--
E
rn -EBUSY;
> +
> +/* there should be better no IOs in flight when we truncate the
> device */
Grammar. Not sure what you were trying to say, but this might be better:
/* There should not be any in-flight IOs when we truncate the device. */
--
Eric Blake eblake redhat com+1
ry data during backup.
> +
> +=Archive format requirements=
> +
> +The basic requirement for such new format is that we can store image
> +date 'out of order'. It is also very likely that we have less than 256
> +drives/images per VM, and we want to be able to store VM configuration
> +files.
> +
> +We have defined a very simply format with those properties, see:
> +
> +docs/specs/vma_spec.txt
This file should be part of the same patch that first mentions it.
> +
> +Please let us know if you know an existing format which provides the
> +same functionality.
> +
> +
>
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
hat qemu_aio_flush() returns.
> + * (without, VM does not reboot)
> +* Note: use 1000 instead of 0 (0 priorize this task too much)
s/priorize/prioritize/
> +++ b/backup.h
> @@ -0,0 +1,32 @@
> +/*
> + * QEMU backup related definitions
> + *
> + * Copyright (C) P
be read only.
>
> Signed-off-by: Wenchao Xia
> Reviewed-by: Eric Blake
Per the cover letter, you modified this patch in v6 by changing a return
type from int to bool. In general, a reviewed-by should be dropped if
your changes were non-trivial, to point out that the reviewer needs t
sertions(+), 5 deletions(-)
I checked that all existing callers only used the function in a boolean
context.
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
t QDict *qdict)
> {
> Error *err = NULL;
> SnapshotInfoList *list;
> +const char *device = qdict_get_try_str(qdict, "device");
> +bool has_device = device ? true : false;
The prevailing style appears to be:
bool has_device = !!device;
--
Eric Blake eb
e functions public, shouldn't you fix them to live
in the right space? Everything else in this .h starts with bdrv_ for a
reason.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
refcount.c | 12 +++-
> block/qcow2-snapshot.c | 7 +--
> 2 files changed, 4 insertions(+), 15 deletions(-)
Always fun when a net decrease in code results in better performance.
Series: Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt v
#x27;*devlist': 'str', '*speed': 'int' },
> + 'returns': 'str' }
> +
> +##
> +# @query-backup
> +#
> +# Returns information about current/last backup task.
> +#
> +# Returns: @BackupStatus
> +#
> +# Since: 1.5.0
> +##
> +{ 'command': 'query-backup', 'returns': 'BackupStatus' }
> +
> +##
> +# @backup-cancel
> +#
> +# Cancel the current executing backup process.
> +#
> +# Returns: nothing on success
> +#
> +# Notes: This command succeeds even if there is no backup process running.
> +#
> +# Since: 1.5.0
> +##
> +{ 'command': 'backup-cancel' }
> +
> +++ b/qmp-commands.hx
> @@ -889,6 +889,18 @@ EQMP
> },
>
> {
> +.name = "backup",
> +.args_type =
> "backupfile:s,format:s?,config-filename:F?,speed:o?,devlist:s?",
> +.mhandler.cmd_new = qmp_marshal_input_backup,
> +},
> +
> +{
> +.name = "backup_cancel",
This doesn't match the spelling in the .json file.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
A check with bdrv_can_read_snapshot(), was done before collecting
> snapshot info.
>
> Signed-off-by: Wenchao Xia
> Reviewed-by: Eric Blake
Again, your cover letter says you heavily touched this patch; leaving my
Reviewed-by intact is misleading.
> +++ b/block.c
> @@ -2649,6 +2649,
on backing images, false or
omitted to show just the top image of a block device
> +#
> +# Returns: a list of @DeviceImageInfo describing each virtual block device
> +#
> +# Since: 1.5
> +##
> +{ 'command': 'query-images',
> + 'data': { '*device
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; ---
> block.c | 43 +++
> include/block/block.h |2 +-
> savevm.c | 10 +-
> 3 files changed, 41 insertions(+), 14 deletions(-)
Reviewed-by: Eric Blake
--
Eric Blake eblake redhat com+1-919-3
to get at all of that information, and not be
limited to just the information on the internal snapshots in top.qcow2.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 02/20/2013 03:57 PM, Eric Blake wrote:
> On 02/18/2013 03:46 PM, Wenchao Xia wrote:
>> Hi, Eric
>> About the interface,there is actually requirement to know internal
>> snapshots in an image of a backing file, so I think the API should be
>> improved a
* @plength: pointer to receive the length that all have the same status as
> + * *pstatus.
> + *
> + * Note: after return, start+*plength may have the same status as
> + * start+*plength-1.
> + */
> +QEMU_DLL_PUBLIC
> +int qb_check_allocation(QBlockImage *qbi,
&g
rver Solutions
It's 2013.
> +
> +#define BITS_PER_LONG (sizeof(unsigned long) * 8)
8 is a magic number; you should be using CHAR_BIT from limits.h instead.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
nsion,
and still have clean documentation of what the extensions are, vs. how
an older version will behave. The initial implementation source code
might be easy to read, but that condition is not guaranteed to last.
If reading the source code to determine the header format is as easy as
you
t32_t version;
> +unsigned char uuid[16];
> +int64_t ctime;
> +unsigned char md5sum[16];
> +
> +uint32_t blob_buffer_offset;
> +uint32_t blob_buffer_size;
> +uint32_t header_size;
> +
> +unsigned char reserved[1984];
> +
> +uint32_t config_name
at uses the compiler to enforce that struct foo is exactly 256 bytes,
because the compiler will fail to create an array of negative size if
struct foo ever ends up with a different size.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
a better interface
>> qemu_strerror() that guarantees thread-safety when converting errno to a
>> string?
>
> I thought I can use g_strerror?
g_strerror is not in use in the current qemu code base; but maybe it is
indeed the right alternative? I'm asking because I'm
noticed that this was broken.
Well, libvirt just entered freeze for 1.0.3. I think the best course of
action on libvirt's side is to patch 1.0.3 to flat-out reject any cpumap
that cannot be represented in a syntax understood by qemu 1.4; then a
future libvirt can re-add support fo
p to mark this with #optional, and document the default
state if the parameter is omitted.
> +#
> +# Since: 1.5
> +##
> +{ 'type': 'ChardevStdio', 'data': { '*signal' : 'bool' } }
> +
> +##
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ock/block.h |4 ++
> qemu-img.c| 81
> -
> 3 files changed, 85 insertions(+), 81 deletions(-)
I have reviewed that this is a straight code motion patch, so if you
have to respin the series for other reasons, and all that changes on
this patch is adding a proper commit message
gt; + device_name);
> +}
> +} else {
> +monitor_printf(mon, "Device '%s' have no correspond image
> now.\n",
s/have no correspond image now/has no corresponding image/
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
so work.
>
> The implementation may still restrict to a single backup, of course.
The initial implementation, obviously. The point is that the API should
allow for growth, even if the initial implementation doesn't support
everything that the API allows.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
t; This isn't a state. Suggest to call it @save_vmstate or similar.
Actually, @save-vmstate (prefer '-' over '_' in QMP)
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
in the preferred syntax. Qemu should feel free
to implement it correctly, rather than trying to cater to older (broken)
libvirt's abuse of comma.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
++-
> hw/xen_disk.c | 2 +-
> include/block/block.h | 4 +--
> include/block/block_int.h | 3 +-
> include/qapi/qmp/qdict.h | 2 ++
> qemu-img.c| 6 ++--
> qemu-io.c | 2 +-
> qemu-nbd.c
finally ever make in QMP) needs to act on all devices at once; reverting
to id '1' or name 'foo' doesn't make much sense, because those are not
consistent. A management app might be able to piece things together
through a more complicated disk-at-a-time transaction, at a
le
it here (even if our handling is reporting an error about invalid usage)?
> +}
> +qdict_del(options, "lazy_refcounts");
> +} else {
> +s->use_lazy_refcounts =
> +(s->compatible_features & QCOW2_COMPAT_LAZY_REFCOUNTS);
> + }
What hap
t be
> a problem.
Concur - send-key is listed as 'Since: 1.3.0', so we are free to change
it however we see fit.
Reviewed-by: Eric Blake
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
VirtIOBlock *s = req->dev;
>
> +snprintf(serial, DEFAULT_VIRTIO_BLK_SERIAL_LEN,
> + "VD%05d", s->drive_serial);
So if I cycle through hotplug and unplug for 10 times, my string
will now be silently truncated, and I can end up with disks with shared
seria
ault serial
> number if user does not specify it.
>
> [v1->v2 address comments from Eric Blake]:
> fix spell errors in patch description
> decrease drive_serial in virtio_blk_exit as well
Typically, patch changelogs belong...
>
> Signed-off-by: Dave Young
> ---
...afte
it a bit odd to name this 'hex', even though it works to
do 'hex':32 as a synonym for 'hex':0x20? Should we instead name it
'value', since we don't care in what base the value was represented,
only that JSON was able to decode the base into a value?
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
> qapi-schema.json | 19 +++---
> 3 files changed, 83 insertions(+), 54 deletions(-)
Series:
Reviewed-by: Eric Blake
However, I have a bikeshed question on patch 2 that may require a respin.
--
Eric Blake ebl...@redhat.com
The rest of the file doesn't have TABs, so you
need to respin this.
> +# IMPORTANT: this option can make QEMU allocates several gigabytes
s/allocates/allocate/
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
already declared that
HMP can change for better UI even if it is not back-compatible.
Reviewed-by: Eric Blake
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
don't care in what base the value was represented,
>>> only that JSON was able to decode the base into a value?
>>
>> Yes, that's a good point. I'll respin.
>
> Actually, this also has drawbacks:
>
> keylist->value->kind = KEY_VALUE_KIND_VA
fd = s->fd;
> size_t writen_size;
While you are here, s/writen/written/ in the local variable.
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 09/21/2012 12:18 PM, Eric Blake wrote:
>> Any better ideas?
>
> Maybe s/hex/number/, as in:
>
> JSON: 'number':0x20
>
> C code: keylist->value->number = number;
>
> that is, you are passing the value either as a keycode name, or as a number.
Or
tries to raise the bandwidth as high as
possible in cases where the user did not request a particular bandwidth.
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
+#
> +##
> +{ 'type': 'MemHpInfo',
> + 'data': {'dimm': 'str', 'request': 'str', 'result': 'str'} }
Should 'result' be a bool (true for success, false for still pending) or
an enum
gt; +.name = "memory-total",
We're split on HMP naming conventions ('-' vs. '_'); we have 'show
migrate_capabilities' but 'show block-jobs'. Oh well, that's life.
Reviewed-by: Eric Blake
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
gnificantly impacting live migration compatibility.
> +
> +### qElsewhere fields
> +
> +In some cases state is saved-off when serializing a seperate device
s/seperate/separate/
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
exer.py
> @@ -0,0 +1,306 @@
> +#
> +# QEMU Lexer Library
> +#
> +# Copyright IBM, Corp. 2012
> +#
> +# Authors:
> +# Anthony Liguori
> +# Michael Roth
> +#
> +# This work is licensed under the terms of the GNU GPLv2.
Any specific reason this is not GPLv2+?
--
; new file mode 100644
> index 000..fe155f7
> --- /dev/null
> +++ b/scripts/qidl_parser.py
> @@ -0,0 +1,262 @@
> +#
> +# QEMU IDL Parser
> +#
> +# Copyright IBM, Corp. 2012
> +#
> +# Authors:
> +# Anthony Liguori
> +# Michael Roth
> +#
> +# This wor
nu_printf__
> +# endif
Don't you need to make this conditional on the gcc version? That is,
gcc 4.4 added __gnu_printf__, but gcc 4.3 understands only __printf__
and attempts to pass __gnu_printf__ will cause compilation failure.
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
} }
properly fails, since 'mid' is not a backing file of 'backing'. I saw
code in patch 1/7 that bdrv_drop_intermediate() should detect the
situation, but I'm not confident enough in my reading of patch 2/7 to
know if that detection point was early enough, or whether
al_err);
Is it valid to make a no-op call, such as:
{ "execute":"block-commit", "arguments":{
"device":"drive0", "top":"base", "base":"base" }}
If so, should we do an early exit here, rather than temporarily chan
On 09/25/2012 12:58 PM, Jeff Cody wrote:
> On 09/25/2012 02:12 PM, Eric Blake wrote:
>> On 09/25/2012 10:29 AM, Jeff Cody wrote:
>>> This adds the live commit coroutine. This iteration focuses on the
>>> commit only below the active layer, and not the active layer itse
tion: Any reason patch 1/7 adds two helper functions, and
patch 4/7 adds a third? Perhaps it would be worth squashing all three
into one commit, or else having three commits, one per helper? But
there's no point in repainting things now, so:
Reviewed-by: Eric Blake
--
Eric Blake ebl...@
per second
> +#
> +# Returns: Nothing on success
> +# If commit or stream is already active on this device, DeviceInUse
> +# If @device does not exist, DeviceNotFound
> +# If image commit is not supported by this device, NotSupported
> +# If
specify 'top' to avoid an error when not committing the active
layer is not as bad as getting the defaults wrong for when we do add
support for committing the active layer.
> ---
>
> I can certainly revert back to having the active layer be the top, if
> that is the preference.
G
e maximum progress value
> #
> +# @paused: whether the job is paused (since 1.2)
1.3
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
# @force: #optional whether to allow cancellation of a paused job (default
> false)
> +#
Do we need (since 1.3) designation on this argument?
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
edding - should the order of the docs match the order of the
'data' array (that is, should 'enospc' be last in both places)?
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
nding event queue, it would be nice to know
whether the error stemmed from a 'stream', 'mirror', or (when combined
with Jeff's patches) 'commit' job.
> +++ b/qapi-schema.json
> @@ -1127,11 +1127,14 @@
> #
> # @speed: the rate limit, bytes per second
> #
> +# @io-status: the status of the job (since 1.2)
1.3
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
On 09/26/2012 01:10 PM, Eric Blake wrote:
>> +Emitted when a block job encounters an error.
>> +
>> +Data:
>> +
>> +- "device": device name (json-string)
>> +- "operation": I/O operation (json-string, "read" or "write&
-
> hmp.c| 8 ++--
> qapi-schema.json | 32
> 4 files changed, 33 insertions(+), 30 deletions(-)
>
Series:
Reviewed-by: Eric Blake
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
+# @on-error: #optional the action to take on an error (default report).
> +#'stop' and 'enospc' can only be used if the block device
> +#supports io-status (see BlockInfo). Since 1.2.
1.3
--
Eric Blake ebl...@redhat.com+1-919-301-326
put: index_from_key(): drop unused code
>
> hmp.c| 43 +---
> input.c | 75
>
> qapi-schema.json | 20 ---
> 3 files changed, 84 insertions(+), 54 deletions(-)
&
n": "write",
> + "action": "stop" },
You didn't document operation and action; are you missing documentation
for this event, or are they bad copy-and-paste in the example?
Again, libvirt would really like to have 'type':'
erate on (json-string)
> +- "target": name of new image file (json-string)
> +- "format": format of new image (json-string, optional)
> +- "mode": how an image file should be created into the target
> + file/device (NewImageMode, optional, default 'a
sts/040
> @@ -0,0 +1,353 @@
> +#!/usr/bin/env python
> +#
> +# Tests for image mirroring.
> +#
> +# Copyright (C) 2012 IBM Corp.
> +#
While you are copying material from IBM, I think you have enough
original content to also claim a Red Hat copyright on this file.
--
Eric Blake e
At any rate, this report spurred me to
finally file an enhancement request against POSIX:
http://www.austingroupbugs.net/view.php?id=617
Someday in the future, we might just get ffsl standardized. :)
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http:/
misleading. Don't you mean:
Thus, we do this instead of (HBITMAP_LOG_MAX_SIZE + BITS_PER_LEVEL -
1)/BITS_PER_LEVEL.
(aka ceil(1.0*HBITMAP_LOG_MAX_SIZE / BITS_PER_LEVEL))
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
0x6803f857 - Feature name table
> +0xCD8E819B - Deduplication table offset and length
Missing a description of what this looks like (see the section on '==
Feature name table ==' for comparison).
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
{
> +.name = "block-commit",
> +.args_type = "device:B,base:s?,top:s?,speed:o?",
s/top:s?/top:s/ now that top is mandatory.
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
e base image of a chain
> QAPI: add command for live block commit, 'block-commit'
> qemu-iotests: add initial tests for live block commit
Series:
Reviewed-by: Eric Blake
I found one minor problem in 5/6, and wonder if that can be fixed by
whoever sends the pull request without
at the
> + * packet is large enough to contain the returned ip header.
> * Only then can we do the check to see if 64 bits of packet
> * data have been returned, since we need to check the returned
> * ip header length.
>
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Lib
is
exactly 0). [Serves me right for not realizing this aspect of JSON when
I did my review earlier.]
I don't think this invalidates the QMP (libvirt already sends decimal,
and wasn't planning on sending hex), but you DO have a point that:
>
> qmp-commands.hx also needs to be upda
new event, possibly as an enum rather than freeform string.
But since we agreed to defer that to a followup patch, it does not hold
up this series. I also did one last sanity read through the patches.
Series:
Reviewed-by: Eric Blake
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Lib
e to type
> 2. We loose file path autocompletion
If it's not too late...
s/loose/lose/
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
|2 +-
> block/vmdk.c |6 +++---
> block/vvfat.c |2 +-
> 4 files changed, 6 insertions(+), 6 deletions(-)
Yay - I've previously suggested this, so it's nice to see it fixed.
Reviewed-by: Eric Blake
--
Eric Blake ebl...@redhat.com+1-919-301-
y for symmetry in the output? But it's
bike-shedding, so I'll live with whatever works (that is, I'm not
requesting a v3 on this patch).
> +##
> +# @SocketAddress
> +#
> +# Captures the address of a socket, which could also be a named file
> descriptor
> +#
> +# Since 1
#optional
#optional should come right after @writable:
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
",", &sptr);
>
> This would break build since strtok_r() isn't available on Mingw, it's
Correct. Microsoft is stuck in the past when it comes to standard
conformance.
> not Posix either.
Huh? http://pubs.opengroup.org/onlinepubs/9699919799/functions/strtok_r.htm
On 10/02/2012 03:00 AM, Paolo Bonzini wrote:
> Il 02/10/2012 01:56, Eric Blake ha scritto:
>> On 10/01/2012 08:52 AM, Paolo Bonzini wrote:
>>> Signed-off-by: Paolo Bonzini
>>> ---
>>> +#
>>> +# @port: port part of the address, or lowest port if @to
plicate': 'int', 'normal': 'int', 'normal-bytes': 'int' } }
> + 'duplicate': 'int', 'normal': 'int', 'normal-bytes': 'int',
> + 'dirty-pages-rate' : 'int' } }
Consistent formatting:
s/ :/:/
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
ened (json-int)
I think the convention is to indent the second line to the same column
as the first character after the colon. But unless it affects the
appearance of any actual generated documentation, I don't think it's
worth a respin for two fewer spaces.
--
Eric Blake ebl...
dirty-log
> +#
> +# Enable or disable the global dirty log mode.
> +#
> +# @enable: true to enable, false to disable.
> +#
> +# Returns: nothing
> +#
> +# Since: 1.2
1.3
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
value of
‘setfsuid’, declared with attribute warn_unused_result
[-Werror=unused-result]
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
tirely opposed to your proposed v1 approach of minimal diff,
and saving formatting for a separate patch; but now that you've done the
work, v2 is also nice and relatively easy to follow. Yes, I also hope
this unblocks the patch.
>
> NB: v1 of this patch got a:
> Signed-off-by: Da
gt; CAP_DAC_OVERRIDE,
> };
>
> -setfsgid(gid);
> -setfsuid(uid);
> -
> +if (setfsgid(gid) < 0) {
> +return -errno;
> +}
> +if (setfsuid(uid) < 0) {
> +return -errno;
> +}
At any rate, this silences the compiler wa
se (good) or into /dir2/base (swapped arguments)?
Fortunately, it looks like if I have 'base' <- 'snap1'(/path/to/base) <-
'snap2'(/path/to/snap1) <- 'active'(/path/to/snap2), then things work
for committing snap2 into snap1, when I specify absolute f
On 10/05/2012 11:29 AM, Eric Blake wrote:
>
> which uses the absolute file name base->filename and fails to find base,
> making it impossible to commit into a base file that was referenced via
> relative name in the chain. I think that bdrv_find_backing_image needs
> t
ion that we already support from
the monitor.
Besides, my approach will let me pass in an O_RDONLY fd on 24 and O_RDWR
on 25 into the same set, whereas your approach creates a new set per fd,
so I can't add new fds to the set until the monitor is up and running.
--
Eric Blake ebl
at we need to nail down the command line
implementation to something that is easy enough to use, before coding up
something that locks us in to bad design.
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
to get
the O_RDONLY fd returned, even if it comes later in the set, since it is
nominally safer.
That is, I wonder if this should really be implemented as an iteration
over the set that selects the best possible match, where more than one
fd can match, but at the end of the iteration, we have the cl
hink the command line can be slightly shorter with:
-add-fd fd=24,set=1,opaque=...
with no real loss in information (that is, s/fdset-id/set/), since our
command lines are already quite long. But going longhand to match QMP
doesn't hurt my feelings (libvirt will be automating this all anyway, so
.kernel_cmdline
(which happened to be .initrd_filename). That is, gcc didn't complain
because it worked by sheer dumb luck.
--
Eric Blake ebl...@redhat.com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
401 - 500 of 23401 matches
Mail list logo
