offlist. Wow this looks awesome, thanks for tackling it! I'll give the
cover letter a response tomorrow when I've had time to digest it
On 12/9/19 6:15 PM, Daniel Henrique Barboza wrote:
> (series based on master commit 97cafa610ecf5)
>
> This work was proposed by Cole in [1]. This is Cole's
r disk of type nvme this element
> +specifies the PCI address of the host NVMe
> +controller.
> +Since 5.6.0
Same
> +
>
>
>
> -For a "file" or "volume" disk type which represents a cdrom
t; ---
> src/qemu/qemu_migration.c | 25 -
> 1 file changed, 20 insertions(+), 5 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
empa
> ---
> src/libvirt_private.syms | 1 +
> src/util/virpci.c| 14 ++
> src/util/virpci.h| 4
> 3 files changed, 19 insertions(+)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
nged, 37 insertions(+), 21 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
rtions(+), 15 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
: Michal Privoznik
> ---
> src/libvirt_private.syms | 1 +
> src/qemu/qemu_domain.c | 11 +--
> src/util/virpci.c| 15 +++
> src/util/virpci.h| 1 +
> 4 files changed, 18 insertions(+), 10 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
> 2 files changed, 29 insertions(+), 46 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
es than just
> virDomainHostdev.
>
> Signed-off-by: Michal Privoznik
> ---
> src/qemu/qemu_cgroup.c | 52 --
> src/qemu/qemu_domain.c | 98 +-
> src/qemu/qemu_domain.h | 9 ++--
> 3 files changed, 42 insertions(+),
znik
> ---
> src/qemu/qemu_domain.c | 18 ++
> 1 file changed, 6 insertions(+), 12 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
evCgroup(virDomainObjPtr vm,
> goto cleanup;
> }
>
Maybe just add a comment here explaining that, at VM startup, this can
lead to /dev/vfio/vfio being added to the cgroup multiple times, but
that has never shown to be a problem.
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
Adjust virLXCDriverGetCapabilities to fill in driver->caps if it is
empty, regardless of the passed 'refresh' value. This matches the
pattern used in virQEMUDriverGetCapabilities
This fixes LXC XML startup parsing for me
Signed-off-by: Cole Robinson
---
v2:
On 12/4/19 9:20 AM, Daniel P. Berrangé wrote:
> The XML parser currently calls virCapabilitiesDomainDataLookup during
> parsing to find the domain capabilities matching the triple
>
> (virt type, os type, arch)
>
> This is, however, bogus with the QEMU driver as it assumes that there
> is an
On 12/9/19 1:15 PM, Cole Robinson wrote:
> Adjust virLXCDriverGetCapabilities to fill in driver->caps if it has
> never been set, regardless of the passed 'refresh' boolean.
>
> This fixes LXC XML startup parsing for me
>
> Signed-off-by: Cole Robinson
> ---
> src
Adjust virLXCDriverGetCapabilities to fill in driver->caps if it has
never been set, regardless of the passed 'refresh' boolean.
This fixes LXC XML startup parsing for me
Signed-off-by: Cole Robinson
---
src/lxc/lxc_conf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/
'
Signed-off-by: Cole Robinson
---
src/qemu/qemu_domain.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
index a15364a71e..6f53e17b6a 100644
--- a/src/qemu/qemu_domain.c
+++ b/src/qemu/qemu_domain.c
@@ -4687,8 +4687,12
Extend configure to pass the detect python binary to C code, and
use it in the test suite, rather than searching PATH
Signed-off-by: Cole Robinson
---
configure.ac | 1 +
tests/testutils.c | 12 +---
2 files changed, 2 insertions(+), 11 deletions(-)
diff --git a/configure.ac b
The path needs to be adjusted for the new script location
Signed-off-by: Cole Robinson
---
tests/testutils.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/testutils.c b/tests/testutils.c
index 9312ff761e..a8215afa42 100644
--- a/tests/testutils.c
+++ b/tests
the comment suggests, but I
didn't investigate it. The patch sidesteps the issue and seems
worth doing anyways
Cole Robinson (2):
tests: fix REGENERATE test-wrap-argv.py usage
tests: use PYTHON detected from configure
configure.ac | 1 +
tests/testutils.c | 14 ++
2 files changed, 3
/to/libvirt.git/run ./setup.py build
Signed-off-by: Cole Robinson
---
src/libvirt-lxc.pc.in | 2 +-
src/libvirt-qemu.pc.in | 2 +-
src/libvirt.pc.in | 6 +++---
3 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/src/libvirt-lxc.pc.in b/src/libvirt-lxc.pc.in
index 27e3de4244
-strcase gnulib module
>
> bootstrap.conf| 1 -
> build-aux/syntax-check.mk | 2 +-
> src/internal.h| 11 +--
> tests/virhashtest.c | 2 +-
> 4 files changed, 7 insertions(+), 9 deletions(-)
>
Reviewed-by: Cole Robinson
- Cole
--
On 11/29/19 4:40 AM, Jidong Xia wrote:
> With this patch users can cold unplug some sound devices.
> use "virsh detach-device vm sound.xml --config" command.
>
> Signed-off-by: Jidong Xia
> ---
> src/conf/domain_conf.c | 48
>
>
ping
On 10/11/19 3:14 PM, Cole Robinson wrote:
> Teach virt-aa-helper how to label a qcow2 data_file, tracked internally
> as externalDataStore. It should be treated the same as its sibling
> disk image
>
> Signed-off-by: Cole Robinson
> ---
> Compiled but not runtime
'cfg' is never initialized here, which causes a crash
later in qemuCheckpointCreateFinalize
Signed-off-by: Cole Robinson
---
src/qemu/qemu_checkpoint.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/qemu/qemu_checkpoint.c b/src/qemu/qemu_checkpoint.c
index d5af6b99bf
On 12/3/19 12:17 PM, Peter Krempa wrote:
> This allows to start and manage the backup job.
>
> Signed-off-by: Peter Krempa
> ---
> po/POTFILES.in | 1 +
> src/qemu/Makefile.inc.am | 2 +
> src/qemu/qemu_backup.c | 941 +++
>
On 11/20/19 9:48 AM, Pavel Hrdina wrote:
> Changes in v2:
> - fixed missing parentheses in patch 01
> - added patch 02 to cover c_isascii
>
With the two issues I pointed out fixed:
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redh
On 11/20/19 9:48 AM, Pavel Hrdina wrote:
> Signed-off-by: Pavel Hrdina
> ---
> src/conf/network_conf.c | 3 +--
> src/qemu/qemu_domain.c | 3 +--
> src/qemu/qemu_qapi.c| 4 +---
> src/util/virconf.c | 2 +-
> src/util/virfile.c | 4 ++--
> 5 files changed, 6 insertions(+), 10
On 11/20/19 9:48 AM, Pavel Hrdina wrote:
> GLib doesn't provide alternative to c_isascii and this is the only usage
> of that macro so define a replacement ourselves.
>
> Signed-off-by: Pavel Hrdina
> ---
> src/util/virkeyfile.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
>
As of 0985a9597bb03 we no longer distribute generated files. mingw
needs to call genprotocol.pl itself, which means it needs rpcgen
and cpp installed.
Signed-off-by: Cole Robinson
---
mingw-libvirt.spec.in | 4
1 file changed, 4 insertions(+)
diff --git a/mingw-libvirt.spec.in b/mingw
n", 1) != 0)
> vah_error(ctl, 1, _("could not set IFS"));
>
g_setenv returns gboolean, FALSE on failure. So these need to be
adjusted to == FALSE, or !g_setenv. virt-aa-helper test is failing after
this
There's a few other instances of '< 0' checks which need to be adjusted
too. With those fixed, for the series:
Reviewed-by: Cole Robinson
But probably for after freeze
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
RN("terminating job '%s' with invalid data", job->name);
> -goto cleanup;
> -}
> -
> if ((job->newstate == QEMU_BLOCKJOB_STATE_COMPLETED ||
> job->newstate == QEMU_BLOCKJOB_STATE_FAILED) &&
> job->state == QEMU_BL
+-
> src/qemu/qemu_blockjob.h | 3 +++
> src/qemu/qemu_domain.c| 13 +-
> src/qemu/qemu_driver.c| 1 +
> .../blockjob-blockdev-in.xml | 1 +
> 5 files changed, 41 insertions(+), 2 d
ic virClassPtr qemuBlockJobDataClass;
>
Need an extra newline here to preserve previous spacing
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
pdate the state to 'QEMU_BLOCKJOB_STATE_COMPLETED' which is probably
> the safest thing to do here.
>
> Signed-off-by: Peter Krempa
> ---
> src/qemu/qemu_blockjob.c | 21 ++---
> 1 file changed, 6 insertions(+), 15 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing
rc < 0)
> qemuBlockJobUnregister(job, vm);
> +else
> +job->reconnected = true;
> continue;
> }
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
rtions(+), 1 deletion(-)
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
EV.
>
> Signed-off-by: Peter Krempa
> ---
> src/qemu/qemu_blockjob.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
d, 3 insertions(+), 3 deletions(-)
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
((job->newstate == QEMU_BLOCKJOB_STATE_COMPLETED ||
> + job->newstate == QEMU_BLOCKJOB_STATE_FAILED) &&
> job->state == QEMU_BLOCKJOB_STATE_ABORTING)
> job->newstate = QEMU_BLOCKJOB_STATE_CANCELLED;
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
| 6 ++
> 2 files changed, 8 insertions(+), 2 deletions(-)
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
a
> ---
> src/qemu/qemu_capabilities.c | 17 +
> src/qemu/qemu_capabilities.h | 2 ++
> 2 files changed, 19 insertions(+)
>
Reviewed-by: Cole Robinson
this made me go look at the case that _BLOCKDEV is disabled. bus='sd',
hmm that's unfortunate. Do you know
On 11/25/19 7:53 AM, Peter Krempa wrote:
> We clear some capabilities here so the lockouts need to be
> re-evaluated.
>
> Signed-off-by: Peter Krempa
> ---
> src/qemu/qemu_process.c | 3 +++
> 1 file changed, 3 insertions(+)
>
Reviewed-by: Cole Robinson
- Cole
--
tions(-)
>
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/25/19 7:53 AM, Peter Krempa wrote:
> Do all post-processing of capabilities in qemuProcessPrepareQEMUCaps.
>
> Signed-off-by: Peter Krempa
> ---
> src/qemu/qemu_process.c | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
Reviewed-by: Cole Robinso
s.
>
> Signed-off-by: Peter Krempa
> ---
> src/qemu/qemu_process.c | 12 +++-
> 1 file changed, 7 insertions(+), 5 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/25/19 7:53 AM, Peter Krempa wrote:
> Start aggregating all capability post-processing code in one place.
>
> The comment was modified while moving it as it was mentioning floppies
> which are no longer clearing the blockdev capability.
>
> Signed-off-by: Peter Krempa
> ---
>
On 11/25/19 9:40 AM, Pavel Mores wrote:
> On Mon, Nov 25, 2019 at 08:54:43AM -0500, Cole Robinson wrote:
>> On 11/25/19 5:54 AM, Pavel Mores wrote:
>>> This new version mostly integrates Cole's comments about the second version.
>>> Refactoring and behaviour change are n
o type selection algorithm
Reviewed-by: Cole Robinson
and pushed now.
patch #3 still had some style issues I pointed out in the previous
review: using 'else' when it isn't necessary, because every block
returns. Better to use the style of qemuDomainDefaultNetModel which
saves having to do any n
On 11/14/19 4:57 PM, Cole Robinson wrote:
> On 10/22/19 11:47 AM, Stefano Garzarella wrote:
>> ruby_libvirt_value_to_int() returns 0 if the optional value is
>> not defined, but in node_cpu_stats() and node_memory_stats()
>> the default value of cpuNum and cellNum is -1.
&
aight revert as additonal patches modified how we store
> the capabilities.
>
> Signed-off-by: Peter Krempa
> ---
> CCing all people in the original conversation:
> CC: Cole Robinson
> CC: Ján Tomko
> CC: Michal Prívozník
>
> src/bhyve/bhyve_
On 11/20/19 9:40 AM, Peter Krempa wrote:
> On Mon, Nov 18, 2019 at 14:43:08 -0500, Cole Robinson wrote:
>> On 11/13/19 11:05 AM, Peter Krempa wrote:
>>> For future extensions of the domain caps it's useful to have a single
>>> point that initializes all capabilities
On 11/20/19 12:25 PM, Jamie Strandboge wrote:
> On Wed, 20 Nov 2019, Cole Robinson wrote:
>
>> On 11/19/19 4:31 PM, Jamie Strandboge wrote:
>>> On Thu, 14 Nov 2019, Christian Ehrhardt wrote:
>>>
>>>> It was mentioned that the pointers in loops like:
&g
On 11/19/19 4:31 PM, Jamie Strandboge wrote:
> On Thu, 14 Nov 2019, Christian Ehrhardt wrote:
>
>> It was mentioned that the pointers in loops like:
>> for (i = 0; i < ctl->def->nserials; i++)
>> if (ctl->def->serials[i] ...
>> will always be !NULL or we would have a much more serious
On 10/25/19 4:28 AM, Patrik Martinsson wrote:
> Hi Tim,
>
> I recently stumbled on the same thing, accidentally shrinking a blockdevice.
>
> I have written a patch for virsh that will force the user to append a
> '--force' flag if shrinking is desired.
>
> The behavior is somewhat still
On 11/18/19 12:02 PM, Peter Krempa wrote:
> Historically we've only supported the as an output-only
> element for domain disks. The documentation states that it may become
> supported on input. To allow management apps detectin once that happens
> add a domain capability which will be asserted if
On 11/13/19 11:05 AM, Peter Krempa wrote:
> For future extensions of the domain caps it's useful to have a single
> point that initializes all capabilities as unsupported by a driver. The
> driver then can enable specific ones.
>
> Signed-off-by: Peter Krempa
> ---
>
On 11/11/19 9:38 AM, Daniel P. Berrangé wrote:
> This series is an effort to reduce the number of different
> languages we use by eliminating most use of perl in favour
> of python.
>
I tested all the new scripts, see individual mails. I didn't really
'review' any of them; since many are just
changed, 3 insertions(+), 3 deletions(-)
> rename {src/hyperv => scripts}/hyperv_wmi_generator.py (100%)
For 20-23:
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
> scripts/genaclperms.py | 123
> 4 files changed, 127 insertions(+), 128 deletions(-)
> delete mode 100755 docs/genaclperms.pl
> create mode 100755 scripts/genaclperms.py
The new script fixes indentation, but otherwise seems to generate
identical aclperms.htmlinc
Tested-by: Col
o the perl script (modulo expected timestamp
change)
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/11/19 9:38 AM, Daniel P. Berrangé wrote:
> As part of an goal to eliminate Perl from libvirt build tools,
> rewrite the check-file-access.pl tool in Python.
>
> This was a straight conversion, manually going line-by-line to
> change the syntax from Perl to Python. Thus the overall structure
wrapping in $file",
> + file=sys.stderr)
> +print("Use test-wrap-argv.py to wrap test data files",
> + file=sys.stderr)
test-wrap-arg reference here
> +return False
> +else:
> +with open(filename, "w") as fh:
> +for line in fixed:
> +print(line, file=fh, end='')
> +
> +return True
> +
> +
> +parser = argparse.ArgumentParser(description='Test arg line wrapper')
test-wrap-arg reference here, should be group-qemu-caps
Otherwise it seems to work as expected
Tested-by: Cole Robinson
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/11/19 9:38 AM, Daniel P. Berrangé wrote:
> As part of an goal to eliminate Perl from libvirt build tools,
> rewrite the test-wrap-argv.pl tool in Python.
>
> This was a straight conversion, manually going line-by-line to
> change the syntax from Perl to Python. Thus the overall structure
>
src/Makefile.am | 98 --
> 4 files changed, 155 insertions(+), 83 deletions(-)
> create mode 100644 scripts/check-remote-protocol.py
I verified the script detected a difference in src/qemu_protocol-structs
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
> src/access/genpolkit.pl| 119
> 4 files changed, 126 insertions(+), 122 deletions(-)
> create mode 100755 scripts/genpolkit.py
> delete mode 100755 src/access/genpolkit.pl
>
Generates identical org.libvirt.api.policy for me
Tested-by: Cole Robinson
- Cole
--
li
ower()):
> +print("%s:%d Mismatch check 'vir%sEnsureACL'" +
> + "for function '%s'" %
> + (filename, lineno, func, maybefunc),
> + file=sys.stderr)
> +errs = True
> +acls[maybefunc] = True
> +elif checkacl:
> +# Record the fact that maybefunc contains an
> +# ACL filter call, and make sure it is the right call!
> +func = checkacl.group(1)
> +if func.startswith("vir"):
> +func = func[3:]
> +
> +if maybefunc is None:
> +print("%s:%d Unexpected check '%s' outside function"
> %
> + (filename, lineno, func), file=sys.stderr)
> +errs = True
> +else:
> +if not maybefunc.lower().endswith(func.lower()):
> +print("%s:%d Mismatch check 'vir%sEnsureACL' " +
The string here should be vir%sCheckACL
After that, and the flake8 fixes, I could trigger the two 'mismatch
check' and 'Missing ACL' errors in qemu_driver.c, which are the
important ones
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
the intent is to match lines like:
.connectSupportsFeature = qemuConnectSupportsFeature, /* 0.5.0 */
Then there needs to be something between '?' and '/'. But this script
could still be useful for sanitizing virStateDriver structs too, which
don't have those ending comments, so maybe just drop the / entirely.
With that and the pep8 issues fixed, all the errors trigger as expected
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/15/19 10:58 AM, Jonathon Jongsma wrote:
> Signed-off-by: Jonathon Jongsma
> ---
> v2:
> Made changes suggested by Andrea. I'll need somebody to push this for me ;)
I pushed it with Andrea's RB tag
Thanks,
Cole
--
libvir-list mailing list
libvir-list@redhat.com
"vir")
> +
> +if sym not in symbols:
> +print("Driver method name %s doesn't match public API" %
> + drv)
Missing status = 1 here
> +continue
> +
> +m = re.search(r
erall structure
> of the file and approach is the same.
>
> Signed-off-by: Daniel P. Berrangé
libvirt_functions.stp looks the same before and after
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
erall structure
> of the file and approach is the same.
>
> Signed-off-by: Daniel P. Berrangé
I verified *.stp generated output is the same across old and new versions
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
erall structure
> of the file and approach is the same.
I verified this catches a bogus symbol name in libvirt_private.syms
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
Makefile.inc.am | 2 +-
> src/check-symsorting.pl | 106
> 5 files changed, 121 insertions(+), 109 deletions(-)
> create mode 100755 scripts/check-symsorting.py
> delete mode 100755 src/check-symsorting.pl
I verified this catches out of order libv
ot sure if the second one at the end of the
file is even triggerable due to compile time protections, but it's still
safe to have
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
-
> scripts/header-ifdef.py | 231 ++
> 4 files changed, 234 insertions(+), 185 deletions(-)
> delete mode 100644 build-aux/header-ifdef.pl
> create mode 100644 scripts/header-ifdef.py
I verified it catches all 3 errors described in the top comment
Tested-by: Cole Robinson
- Cole
--
erall structure
> of the file and approach is the same.
>
> Signed-off-by: Daniel P. Berrangé
Verified it catches a single failure case
Tested-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/11/19 9:38 AM, Daniel P. Berrangé wrote:
> As part of an goal to eliminate Perl from libvirt build tools,
> rewrite the check-spacing.pl tool in Python.
>
> This was a straight conversion, manually going line-by-line to
> change the syntax from Perl to Python. Thus the overall structure
>
On 11/11/19 9:38 AM, Daniel P. Berrangé wrote:
> This series is an effort to reduce the number of different
> languages we use by eliminating most use of perl in favour
> of python.
I'm testing the series now. On fedora 31, make syntax-check is showing
some flake8 errors, see attached. I also
On 11/11/19 9:38 AM, Daniel P. Berrangé wrote:
> As part of an goal to eliminate Perl from libvirt build tools,
> rewrite the check-drivername.pl tool in Python.
>
> This was mostly a straight conversion, manually going line-by-line
> to change the syntax from Perl to Python. Thus the overall
> 1 file changed, 40 insertions(+), 35 deletions(-)
>
Reviewed-by: Cole Robinson
And pushed
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
) {
Add a space after 'switch'. 'make syntax-check' will catch this. libvirt
style is typically to not indent the 'case' keyword either, but this
file is inconsistent on that front. With those fixed:
Reviewed-by: Cole Robinson
This matches what is done for the selinux driv
alNetDefValidate. I think that naming is more consistent
rather than invent a new 'runtime' naming scheme, event though runtime
is more semanticly descriptive. (though TBH I wouldn't mind
s/actual/runtime/ in the codebase anyways, the 'actual' naming always
confuses me)
Comment should be fixed ei
but what you describe
makes sense and matches my reading of the code. -1 is
VIR_NODE_CPU_STATS_ALL_CPUS and VIR_NODE_MEMORY_STATS_ALL_CELLS so it
makes sense that would be the default.
Reviewed-by: Cole Robinson
CCing clalance who is the primary ruby-libvirt maintainer, but if he
doesn't get to it
On 10/20/19 11:54 PM, jcfara...@gmail.com wrote:
> From: Julio Faracco
>
I think if you set your gitconfig correctly, you can avoid this 'From:'
showing up. I have:
[sendemail]
from = "Cole Robinson "
[user]
name = Cole Robinson
email = crobi...@redhat.com
>
On 11/14/19 11:40 AM, Cole Robinson wrote:
> On 10/18/19 11:30 AM, Jonathon Jongsma wrote:
>> This is the third version of a patch series that adds support for a boot
>> display for mediated vgpu devices using the 'ramfb' parameter. This version
>> fixes a bunch of tes
On 10/18/19 11:30 AM, Jonathon Jongsma wrote:
> This is the third version of a patch series that adds support for a boot
> display for mediated vgpu devices using the 'ramfb' parameter. This version
> fixes a bunch of test failures that I had inadvertently introduced in the last
> series. It also
1 file changed, 29 insertions(+), 43 deletions(-)
>
I tweaked the patch subjects, to use 'qemu: hotplug:' prefix, the X.c
format isn't typically used. Patch 2 incorrectly mentioned qemu_process,
so I fixed that too
Reviewed-by: Cole Robinson
and pushed
- Cole
--
libvir-list mailing list
li
mplify the if chains in get_files by dropping these checks.
>
> Signed-off-by: Christian Ehrhardt
> ---
> src/security/virt-aa-helper.c | 135 --
> 1 file changed, 63 insertions(+), 72 deletions(-)
>
Reviewed-by: Cole Robinson
- Cole
--
li
On 10/15/19 2:41 AM, Jidong Xia wrote:
> With this patch users can cold unplug some sound devices.
> use "virsh detach-device vm sound.xml --config" command.
>
I reviewed and pushed patch #1
Some comments below
> Signed-off-by: Jidong Xia
> ---
> src/conf/domain_conf.c | 61
>
libxl_migration.c | 14 +-
> 2 files changed, 12 insertions(+), 16 deletions(-)
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
| 35
> src/conf/network_conf.c | 4 +---
> src/qemu/qemu_migration_params.c | 7 +--
> 3 files changed, 15 insertions(+), 31 deletions(-)
>
Reviewed-by: Cole Robinson
And pushed
Thanks,
Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 11/13/19 5:41 PM, Jonathon Jongsma wrote:
> On Wed, 2019-11-13 at 13:54 -0500, Cole Robinson wrote:
>> I pushed the first three patches. Comments below
>>
>> On 10/23/19 1:46 PM, Jonathon Jongsma wrote:
>>> The current code doesn't properly handle errors when pa
buf);
> +return -EINVAL;
> +}
> +
> +*tmp = '\0';
> +
> +if (virStrToDouble(buf, NULL, ) < 0) {
> +virReportError(VIR_ERR_INTERNAL_ERROR,
> + _("Unable to parse sched info value '%s'"),
> + buf);
> +return -EINVAL;
> +}
> +
error message here is a copy+paste mistake. syntax-check says you need
to extend POTFILES.in too. With those fixed:
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
insertions(+), 47 deletions(-)
>
Not runtime tested, but:
Reviewed-by: Cole Robinson
- Cole
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
ev/shm/%s", shmem->name) <
> 0)
> + goto cleanup;
virAsprintf is gone in git, you can use g_strdup_printf, which also
means dropping the error checking. See the other conversions patches
With those changes, for the series:
Reviewed-by: Cole Robi
On 11/13/19 8:22 AM, Martin Kletzander wrote:
> On Mon, Oct 07, 2019 at 02:56:11PM +0200, Pavel Mores wrote:
>>
>> Hi,
>>
>> I'm looking into fixing
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=1668141
>>
>> (as a short summary, if a graphics device is added to XML that has no
>> video
>>
Sorry for the review delay. If you cc me on v3 I'll review it ASAP
This patch is mixing two things: refactoring, and behavior change. This
makes review more difficult. Please put the refactoring first which will
maintain the old behavior, then add functional changes as additional
patch(es).
On
On 10/18/19 2:36 PM, Daniel Henrique Barboza wrote:
> changes in v3:
> - redesigned patch 1 based on Cole Robinson feedback
>
> v2: https://www.redhat.com/archives/libvir-list/2019-September/msg00447.html
> v1: https://www.redhat.com/archives/libvir-list/2019-August/msg00698.html
On 10/23/19 1:46 PM, Jonathon Jongsma wrote:
> In order to differentiate between a configuration that does not specify
> any video acceleration and one that is incorrectly specified, change the
> signature of virDomainVideoAccelDefParseXML() to return an error
> response.
>
> If any of the values
I pushed the first three patches. Comments below
On 10/23/19 1:46 PM, Jonathon Jongsma wrote:
> The current code doesn't properly handle errors when parsing a video
> device's resolution.
>
> This patch changes the parse function signature to return an error
> when we're missing an 'x' or 'y'
201 - 300 of 3789 matches
Mail list logo
