support.
Any comments appreciated,
Thanks!
Joao
Joao Martins (5):
public/xen.h: add flags field to vcpu_time_info
x86/time: implement tsc as clocksource
x86/time: streamline platform time init on plt_init()
x86/time: refactor read_platform_stime()
x86/time: implement PVCLOCK_TSC_ST
On 03/18/2016 08:12 PM, Andrew Cooper wrote:
> On 17/03/16 16:12, Joao Martins wrote:
>> This field has two possible flags (as of latest pvclock ABI
>> shared with KVM).
>>
>> flags: bits in this field indicate extended capabilities
>> coordinated bet
On 03/18/2016 08:21 PM, Andrew Cooper wrote:
> On 17/03/16 16:12, Joao Martins wrote:
>> Introduce support for using TSC as platform time which is the highest
>> resolution time and most performant to get (~20 nsecs). Though there
>> are also several problems associate
On 03/18/2016 08:32 PM, Andrew Cooper wrote:
> On 17/03/16 16:12, Joao Martins wrote:
>> And use to initialize platform time solely for clocksource=tsc,
>> as opposed to initializing platform overflow timer, which would
>> only fire in ~180 years (on 2.2 Ghz Broadwell pro
On 03/18/2016 08:34 PM, Andrew Cooper wrote:
> On 17/03/16 16:12, Joao Martins wrote:
>> To fetch the last read from the clocksource which was used to
>> calculate system_time. In the case of clocksource=tsc we will
>> use it to set tsc_timestamp.
>>
>> Signed
On 03/18/2016 08:58 PM, Andrew Cooper wrote:
> On 17/03/16 16:12, Joao Martins wrote:
>> When using TSC as clocksource we will solely rely on TSC for updating
>> vcpu time infos (pvti). Right now, each vCPU takes the tsc_timestamp at
>> different instants meaning every EPOC
On 03/21/2016 11:43 AM, Andrew Cooper wrote:
> On 21/03/16 11:42, Joao Martins wrote:
>>
>> On 03/18/2016 08:12 PM, Andrew Cooper wrote:
>>> On 17/03/16 16:12, Joao Martins wrote:
>>>> This field has two possible flags (as of latest pvclock ABI
>>>
On 03/21/2016 03:27 PM, Andrew Cooper wrote:
> On 21/03/16 15:10, Jan Beulich wrote:
> On 17.03.16 at 17:12, wrote:
>>> --- a/xen/include/public/xen.h
>>> +++ b/xen/include/public/xen.h
>>> @@ -614,10 +614,14 @@ struct vcpu_time_info {
>>> */
>>> uint32_t tsc_to_system_mul;
>>>
On 03/18/2016 08:21 PM, Andrew Cooper wrote:
> On 17/03/16 16:12, Joao Martins wrote:
>> Introduce support for using TSC as platform time which is the highest
>> resolution time and most performant to get (~20 nsecs). Though there
>> are also several problems associate
On 03/22/2016 12:46 PM, Jan Beulich wrote:
>>>> On 22.03.16 at 13:41, wrote:
>
>>
>> On 03/18/2016 08:21 PM, Andrew Cooper wrote:
>>> On 17/03/16 16:12, Joao Martins wrote:
>>>> Introduce support for using TSC as platform time which is the hig
On 03/22/2016 04:02 PM, Jan Beulich wrote:
>>>> On 22.03.16 at 16:51, wrote:
>
>>
>> On 03/22/2016 12:46 PM, Jan Beulich wrote:
>>>>>> On 22.03.16 at 13:41, wrote:
>>>
>>>>
>>>> On 03/18/2016 08:21 PM, Andrew Coope
On 03/23/2016 07:28 AM, Jan Beulich wrote:
On 22.03.16 at 21:40, wrote:
>> On 03/22/2016 04:02 PM, Jan Beulich wrote:
>> On 22.03.16 at 16:51, wrote:
On 03/22/2016 12:46 PM, Jan Beulich wrote:
On 22.03.16 at 13:41, wrote:
>>>
>> I found out one issue in the tsc c
To fetch the last read from the clocksource which was used to
calculate system_time. In the case of clocksource=tsc we will
use it to set tsc_timestamp.
Signed-off-by: Joao Martins
Reviewed-by: Andrew Cooper
---
Cc: Keir Fraser
Cc: Jan Beulich
Cc: Andrew Cooper
Changes since v1:
- Add
And use to initialize platform time solely for clocksource=tsc,
as opposed to initializing platform overflow timer, which would
only fire in ~180 years (on 2.2 Ghz Broadwell processor).
Signed-off-by: Joao Martins
Reviewed-by: Andrew Cooper
---
Cc: Keir Fraser
Cc: Jan Beulich
Cc: Andrew
host
| |
-
Signed-off-by: Joao Martins
---
Cc: Ian Jackson
Cc: Jan Beulich
Cc: Keir Fraser
Cc: Tim Deegan
Changes since v1:
- flags and pad are both uint8_t.
- fix indentation with the other fields in the struct.
---
xen/in
arp < 100 ns. The warps still exist though but
are only related to cross CPU calibration (being how much it takes to
rendezvous with master), in which a later patch in this series aims to
solve.
Signed-off-by: Joao Martins
---
Cc: Keir Fraser
Cc: Jan Beulich
Cc: Andrew Cooper
Changes si
though still requiring PVCLOCK_TSC_STABLE_BIT flag
support.
Thanks!
Joao
[0]
http://bugs.xenproject.org/xen/mid/%3c1458231136-13457-1-git-send-email-joao.m.mart...@oracle.com%3E
Joao Martins (6):
public/xen.h: add flags field to vcpu_time_info
x86/time: refactor init_platform_time()
x86/time: implement t
wly hotplugged CPU may not
satisfy the condition of having all TSCs synchronized.
Signed-off-by: Joao Martins
---
Cc: Keir Fraser
Cc: Jan Beulich
Cc: Andrew Cooper
Perhaps "cpuhotplugsafe" would be a better name, since potentially
hardware could guarantee TSCs are synchronized on h
And accomodate platform time source initialization in
try_platform_time(). This is a preparatory patch for deferring
TSC clocksource initialization to the stage where all CPUS are
up (verify_tsc_reliability init call).
Signed-off-by: Joao Martins
---
Cc: Keir Fraser
Cc: Jan Beulich
Cc: Andrew
On 03/29/2016 06:39 PM, Konrad Rzeszutek Wilk wrote:
>> static void tsc_check_slave(void *unused)
>> @@ -1437,6 +1515,20 @@ static int __init verify_tsc_reliability(void)
>> }
>> }
>>
>> +if ( !strcmp(opt_clocksource, "tsc") )
>
> Pls also update xen-command-line.markdown
>
On 03/30/2016 04:49 PM, Ian Jackson wrote:
> Joao Martins writes ("[PATCH v2 1/6] public/xen.h: add flags field to
> vcpu_time_info"):
>> This field has two possible flags (as of latest pvclock ABI
>> shared with KVM).
>>
>> flags: bits in this field indic
gt;> 0 | 24 | multiple cpus are guaranteed to
>> | | be monotonic
>> -
>> | | guest vcpu has been paused by
>> 1 | N/A | the host
>> | |
>&
On 04/01/2016 05:10 PM, Konrad Rzeszutek Wilk wrote:
> On Tue, Mar 29, 2016 at 02:44:07PM +0100, Joao Martins wrote:
>> And accomodate platform time source initialization in
>> try_platform_time(). This is a preparatory patch for deferring
>> TSC clocksource initialization t
[snip]
>> diff --git a/xen/arch/x86/time.c b/xen/arch/x86/time.c
>> index ed4ed24..2602dda 100644
>> --- a/xen/arch/x86/time.c
>> +++ b/xen/arch/x86/time.c
>> @@ -432,6 +432,63 @@ uint64_t ns_to_acpi_pm_tick(uint64_t ns)
>> }
>>
>> /
On 04/01/2016 07:45 PM, Konrad Rzeszutek Wilk wrote:
> On Fri, Apr 01, 2016 at 07:38:18PM +0100, Joao Martins wrote:
>> [snip]
>>
>>>> diff --git a/xen/arch/x86/time.c b/xen/arch/x86/time.c
>>>> index ed4ed24..2602dda 100644
>>>> --- a
On 04/05/2016 11:09 AM, Jan Beulich wrote:
On 29.03.16 at 15:44, wrote:
>> --- a/xen/arch/x86/time.c
>> +++ b/xen/arch/x86/time.c
>> @@ -533,6 +533,30 @@ static void resume_platform_timer(void)
>> plt_stamp = plt_src.read_counter();
>> }
>>
>> +static int __init try_platform_timer(str
On 04/05/2016 11:16 AM, Jan Beulich wrote:
On 29.03.16 at 15:44, wrote:
>> --- a/xen/include/public/xen.h
>> +++ b/xen/include/public/xen.h
>> @@ -614,10 +614,14 @@ struct vcpu_time_info {
>> */
>> uint32_t tsc_to_system_mul;
>> int8_t tsc_shift;
>> -int8_t pad1[3];
On 04/05/2016 11:43 AM, Jan Beulich wrote:
On 29.03.16 at 15:44, wrote:
>> Introduce support for using TSC as platform time which is the highest
>> resolution time and most performant to get (~20 nsecs). Though there
>> are also several problems associated with its usage, and there isn't a
>
On 04/05/2016 12:46 PM, Jan Beulich wrote:
On 29.03.16 at 15:44, wrote:
>> --- a/xen/arch/x86/time.c
>> +++ b/xen/arch/x86/time.c
>> @@ -434,6 +434,7 @@ uint64_t ns_to_acpi_pm_tick(uint64_t ns)
>> /
>> * PLATFORM TIMER 4: TSC
>>
On 04/05/2016 12:52 PM, Jan Beulich wrote:
On 29.03.16 at 15:44, wrote:
>> -static s_time_t read_platform_stime(void)
>> +static s_time_t read_platform_stime(u64 *stamp)
>> {
>> -u64 count;
>> +u64 plt_stamp_counter, count;
>
> "stamp" and "counter" seem kind of redundant.
>
A bi
On 04/05/2016 04:12 PM, Jan Beulich wrote:
On 05.04.16 at 16:56, wrote:
>> On 04/05/2016 11:43 AM, Jan Beulich wrote:
>> On 29.03.16 at 15:44, wrote:
@@ -541,6 +613,10 @@ static int __init try_platform_timer(struct
platform_timesource *pts)
if ( rc <= 0 )
On 04/05/2016 04:26 PM, Jan Beulich wrote:
On 05.04.16 at 17:22, wrote:
>> On 04/05/2016 12:52 PM, Jan Beulich wrote:
>> On 29.03.16 at 15:44, wrote:
-static s_time_t read_platform_stime(void)
+static s_time_t read_platform_stime(u64 *stamp)
{
-u64 count;
On 04/05/2016 04:22 PM, Jan Beulich wrote:
On 05.04.16 at 17:12, wrote:
>> On 04/05/2016 12:46 PM, Jan Beulich wrote:
>> On 29.03.16 at 15:44, wrote:
@@ -516,17 +519,31 @@ static s_time_t __read_platform_stime(u64
platform_time)
return (stime_platform_stamp + scale_d
On 04/05/2016 01:22 PM, Jan Beulich wrote:
On 29.03.16 at 15:44, wrote:
>> --- a/xen/arch/x86/time.c
>> +++ b/xen/arch/x86/time.c
>> @@ -43,6 +43,10 @@
>> static char __initdata opt_clocksource[10];
>> string_param("clocksource", opt_clocksource);
>>
>> +/* opt_nocpuhotplug: Set if CPU ho
>> The main
>> difference I see between both would be the base system time:
>> read_platform_stime
>> uses stime_platform_stamp as base, and computes a difference from the
>> read_counter (i.e. rdtsc() ) with previously saved platform-wide stamp
>> (platform_timer_stamp). get_s_time uses the stime
On 05/04/2016 08:41 AM, osstest service owner wrote:
> flight 93436 libvirt real [real]
> http://logs.test-lab.xenproject.org/osstest/logs/93436/
>
> Regressions :-(
>
> Tests which did not succeed and are blocking,
> including tests which could not be run:
> test-amd64-amd64-libvirt-qemuu-deb
ll be less effort involved
into making netback work with something like zerogrant.
Cheers,
Joao
Joao Martins (1):
public/io/netif.h: add gref mapping control messages
xen/include/public/io/netif.h | 114 ++
1 file changed, 114 insertions(+)
---
% Staging
added are:
- Add grefs to be mapped by backend
- Remove grefs mappings (If they are not in use)
- Get maximum amount of grefs kept mapped.
Signed-off-by: Joao Martins
---
xen/include/public/io/netif.h | 114 ++
1 file changed, 114 insertions(+)
diff
On 09/06/2017 02:49 PM, Paul Durrant wrote:
>> -Original Message-
>> From: Joao Martins [mailto:joao.m.mart...@oracle.com]
>> Sent: 01 September 2017 15:51
>> To: Xen-devel
>> Cc: Wei Liu ; Paul Durrant ;
>> Konrad Rzeszutek Wilk ; Joao Martins
>
Hey!
We wanted to brought up this small proposal regarding the lack of
parameterization on PV devices on Xen.
Currently users don't have a way for enforce and control what
features/queues/etc the backend provides. So far there's only global parameters
on backends, and specs do not mention anythin
[Forgot two important details regarding Xenbus states]
On 09/07/2017 05:53 PM, Joao Martins wrote:
> Hey!
>
> We wanted to brought up this small proposal regarding the lack of
> parameterization on PV devices on Xen.
>
> Currently users don't have a way for enforce and
On 10/03/2017 12:55 PM, Joao Martins wrote:
> Right now there is only a pvclock_pvti_cpu0_va() which is defined
> on kvmclock since:
>
> commit dac16fba6fc5
> ("x86/vdso: Get pvclock data from the vvar VMA instead of the fixmap")
>
> The only user of this inter
whether it is running in a KVM guest.
Signed-off-by: Joao Martins
---
New in v7;
---
drivers/ptp/ptp_kvm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/ptp/ptp_kvm.c b/drivers/ptp/ptp_kvm.c
index 2b1b212c219e..e04d7b2ecb3a 100644
--- a/drivers/ptp/ptp_kvm.c
+++ b/drivers/ptp
peers noticed when such changes happen.
Signed-off-by: Joao Martins
Acked-by: Juergen Gross
Reviewed-by: Konrad Rzeszutek Wilk
Acked-by: Paolo Bonzini
---
Changes since v4:
* Add Paolo's Acked-by
* Add Konrad's Reviewed-by
Changes since v1:
* Add Juergen's Gross Acked-by.
---
M
anks,
Joao
Joao Martins (5):
ptp_kvm: probe for kvm guest availability
x86/pvclock: add setter for pvclock_pvti_cpu0_va
x86/xen/time: set pvclock flags on xen_time_init()
x86/xen/time: setup vcpu 0 time info page
MAINTAINERS: xen, kvm: track pvclock-abi.h changes
M
Specifically check for PVCLOCK_TSC_STABLE_BIT and if this bit is set,
then set it too on pvclock flags. This allows Xen clocksource to use it
and thus speeding up xen_clocksource_read() callers (i.e. sched_clock())
Signed-off-by: Joao Martins
Reviewed-by: Boris Ostrovsky
---
Changes since v5
and moves pvclock_pvti_cpu0_va
to pvclock, which is a more generic place to have it; and would
allow other PV clocksources to use it, such as Xen.
Signed-off-by: Joao Martins
Acked-by: Andy Lutomirski
---
Changes since v1:
* Rebased: the only conflict was that I had move the export
pvclock_pvti_cpu0_va()
vcpu_time_info struct.
Signed-off-by: Joao Martins
Reviewed-by: Juergen Gross
Reviewed-by: Boris Ostrovsky
---
Changes since v6:
* Add Boris RoB
Changes since v5:
* Move xen_setup_vsyscall_time_info within the PVCLOCK_TSC_STABLE_BIT
clause added in predecessor patch.
Changes since v4:
* Remove
On 10/17/2017 04:34 PM, Joao Martins wrote:
> On 10/03/2017 12:55 PM, Joao Martins wrote:
>> Right now there is only a pvclock_pvti_cpu0_va() which is defined
>> on kvmclock since:
>>
>> commit dac16fba6fc5
>> ("x86/vdso: Get pvclock data from the vvar VMA
The proposed directory provides a mechanism for tools to control the
maximum feature set of the device being provisioned by backend.
The parameters/features include offloading features, number of
queues etc.
Signed-off-by: Joao Martins
---
xen/include/public/io/netif.h | 16
1
ned-off-by: Joao Martins
---
tools/libxl/libxl.h | 8
tools/libxl/libxl_9pfs.c | 2 +-
tools/libxl/libxl_console.c | 2 +-
tools/libxl/libxl_device.c | 14 --
tools/libxl/libxl_internal.h | 2 +-
tools/libxl/libxl_nic.c | 13 -
t
The proposed directory provides a mechanism for tools to control the
maximum feature set of the device being provisioned by backends.
Examples include max ring page order, persistent grants, number of
queues etc.
Signed-off-by: Joao Martins
---
xen/include/public/io/blkif.h | 14
ly specific
to how it's done and could possibly be implemented in other PV interfaces. But
there wasn't a protocol agnostic file to put all this, so I went ahead and did
for the two individual io types (block and netif) I am most interested in.
Any comments appreciated :)
Thanks!
Joa
ire-multi-queue-max-queues=1" ]
...
Which would set the number of queues to 2 as opposed to e.g. the global
blkback defined xen_blkback.max_queues parameter.
Signed-off-by: Joao Martins
---
tools/libxl/libxlu_disk_l.l | 42 ++
1 file changed, 42 inser
multi-queue-max-queues=2"]
...
Which would set the number of queues to 2 as opposed to e.g. the global
netback defined xen_netback.max_queues parameter.
Signed-off-by: Joao Martins
---
tools/xl/xl_parse.c | 37 +
tools/xl/xl_parse.h | 2 ++
2 files ch
a key value store storing:
=
xl|libxl are stateless with respect to feature names therefore is up to the
admin to carefully select those. If backend isn't supported therefore the
features won't be overwritten.
Signed-off-by: Joao Martins
---
tools/libxl/libxl.h | 8 +
ed and all values set there are
used in place of the module parameters being used.
Note, however that feature-barrier, feature-flush-support and
feature-discard aren't probed because first two are physical
device dependent and feature-discard already has tunables to
adjust.
Signed-off-b
for 'feature-no-csum-offload' to allow toolstack
to control per device checksum offloading.
Signed-off-by: Joao Martins
---
drivers/net/xen-netback/xenbus.c | 122 +++
1 file changed, 99 insertions(+), 23 deletions(-)
diff --git a/drivers/net/xen-netb
On Mon, Nov 06, 2017 at 10:33:59AM +, Paul Durrant wrote:
> > -Original Message-
> > From: Joao Martins [mailto:joao.m.mart...@oracle.com]
> > Sent: 02 November 2017 18:06
> > To: Xen Development List
> > Cc: Joao Martins ; Konrad Rzeszutek Wil
On 11/07/2017 11:28 AM, Oleksandr Grytsov wrote:
> On Thu, Nov 2, 2017 at 8:06 PM, Joao Martins <mailto:joao.m.mart...@oracle.com>> wrote:
>
> The function libxl__device_generic_add will have an additional
> argument whereby it adds a second set of entries visible
On 11/06/2017 04:09 PM, Paolo Bonzini wrote:
> On 19/10/2017 15:39, Joao Martins wrote:
>> Right now there is only a pvclock_pvti_cpu0_va() which is defined
>> on kvmclock since:
>>
>> commit dac16fba6fc5
>> ("x86/vdso: Get pvclock data from the vvar VMA inste
On 11/08/2017 11:06 AM, Thomas Gleixner wrote:
> On Tue, 7 Nov 2017, Joao Martins wrote:
>> On 11/06/2017 04:09 PM, Paolo Bonzini wrote:
>>> On 19/10/2017 15:39, Joao Martins wrote:
>>>> Right now there is only a pvclock_pvti_cpu0_va() which is defined
>>&g
peers noticed when such changes happen.
Signed-off-by: Joao Martins
Acked-by: Juergen Gross
Reviewed-by: Konrad Rzeszutek Wilk
Acked-by: Paolo Bonzini
---
Changes since v4:
* Add Paolo's Acked-by
* Add Konrad's Reviewed-by
Changes since v1:
* Add Juergen's Gross Acked-by.
---
M
individual patches)
Thanks,
Joao
Joao Martins (5):
ptp_kvm: probe for kvm guest availability
x86/pvclock: add setter for pvclock_pvti_cpu0_va
x86/xen/time: set pvclock flags on xen_time_init()
x86/xen/time: setup vcpu 0 time info page
MAINTAINERS: xen, kvm: track pvclock-abi.h changes
M
Specifically check for PVCLOCK_TSC_STABLE_BIT and if this bit is set,
then set it too on pvclock flags. This allows Xen clocksource to use it
and thus speeding up xen_clocksource_read() callers (i.e. sched_clock())
Signed-off-by: Joao Martins
Reviewed-by: Boris Ostrovsky
---
Changes since v5
.
Signed-off-by: Joao Martins
Acked-by: Radim Krčmář
---
Changes since v7:
* Add Radim's Acked-by
---
drivers/ptp/ptp_kvm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/ptp/ptp_kvm.c b/drivers/ptp/ptp_kvm.c
index 2b1b212c219e..e04d7b2ecb3a 100644
--- a/drivers/ptp/ptp_kvm.c
the setter (pvclock_set_pvti_cpu0_va).
Signed-off-by: Joao Martins
Acked-by: Andy Lutomirski
Acked-by: Paolo Bonzini
Acked-by: Thomas Gleixner
---
Changes since v7:
* Add Paolo Acked-by
(Comments from Thomas Gleixner)
* Rename getter to pvclock_get_pvti_cpu0_va
and fixup its callsites (vdso/vma
vcpu_time_info struct.
Signed-off-by: Joao Martins
Reviewed-by: Juergen Gross
Reviewed-by: Boris Ostrovsky
---
Changes since v5:
* Move xen_setup_vsyscall_time_info within the PVCLOCK_TSC_STABLE_BIT
clause added in predecessor patch.
Changes since v4:
* Remove pvclock_set_flags since predecessor
er batches (i.e. for better throughput with big packets) as it
was prior to the above mentioned commit.
Signed-off-by: Joao Martins
---
drivers/net/xen-netback/common.h| 6 --
drivers/net/xen-netback/interface.c | 25 -
drivers/net/xen-netback/netback.c | 5
On 11/13/2017 10:33 AM, Paul Durrant wrote:
>> -Original Message-
>> From: Joao Martins [mailto:joao.m.mart...@oracle.com]
>> Sent: 10 November 2017 19:35
>> To: net...@vger.kernel.org
>> Cc: Joao Martins ; Wei Liu
>> ; Paul Durrant ; xen-
>> de.
On Mon, Nov 13, 2017 at 11:58:03AM +, Paul Durrant wrote:
> On Mon, Nov 13, 2017 at 11:54:00AM +0000, Joao Martins wrote:
> > On 11/13/2017 10:33 AM, Paul Durrant wrote:
> > > On 11/10/2017 19:35 PM, Joao Martins wrote:
[snip]
> > >> diff --git a/drivers/net/
On Mon, Nov 13, 2017 at 04:39:09PM +, Paul Durrant wrote:
> > -Original Message-
> > From: Joao Martins [mailto:joao.m.mart...@oracle.com]
> > Sent: 13 November 2017 16:34
> > To: Paul Durrant
> > Cc: net...@vger.kernel.org; Wei Liu ; xen-
> > de
Hey Oleksandr,
On 07/31/2017 09:34 AM, Oleksandr Andrushchenko wrote:
> Hi, all!
>
[snip]
>
> Comparison for display use-case
> ===
>
> 1 Number of grant references used
> 1-1 grant references: nr_pages
> 1-2 GNTTABOP_transfer: nr_pages
> 1-3 XENMEM_exchange: not an
On 07/31/2017 11:37 AM, Oleksandr Andrushchenko wrote:
> On 07/31/2017 01:04 PM, Julien Grall wrote:
>> On 31/07/17 10:52, Oleksandr Andrushchenko wrote:
>>> On 07/31/2017 12:47 PM, Julien Grall wrote:
On 31/07/17 10:46, Oleksandr Andrushchenko wrote:
Do you have any example of hardware?
On 07/31/2017 12:41 PM, Oleksandr Andrushchenko wrote:
> Hi, Joao!
>
> On 07/31/2017 02:03 PM, Joao Martins wrote:
>> Hey Oleksandr,
>>
>> On 07/31/2017 09:34 AM, Oleksandr Andrushchenko wrote:
>>> Hi, all!
>>>
On 08/01/2017 07:34 PM, Andrew Cooper wrote:
> On 31/07/2017 20:49, Konrad Rzeszutek Wilk wrote:
>> On Wed, Jul 05, 2017 at 02:22:00PM +0100, Joao Martins wrote:
>>> On 07/05/2017 12:16 PM, Andrew Cooper wrote:
>>>> On 05/07/17 10:46, Joao Martins wrote:
>>>&
Hey Andrew,
On 07/04/2017 03:55 PM, Andrew Cooper wrote:
> Presented herewith is the a plan for the final part of CPUID work, which
> primarily covers better Xen/Toolstack interaction for configuring the guests
> CPUID policy.
>
Really nice write up, a few comments below.
> A PDF version of this
On 07/05/2017 10:46 AM, Joao Martins wrote:
> Hey Andrew,
>
> On 07/04/2017 03:55 PM, Andrew Cooper wrote:
>> Presented herewith is the a plan for the final part of CPUID work, which
>> primarily covers better Xen/Toolstack interaction for configuring the guests
>> CPU
On 07/05/2017 12:16 PM, Andrew Cooper wrote:
> On 05/07/17 10:46, Joao Martins wrote:
>> Hey Andrew,
>>
>> On 07/04/2017 03:55 PM, Andrew Cooper wrote:
>>> Presented herewith is the a plan for the final part of CPUID work, which
>>> primarily covers better Xe
added are:
- Add grefs to be mapped by backend
- Remove grefs mappings (If they are not in use)
- Get maximum amount of grefs kept mapped.
Signed-off-by: Joao Martins
---
v3:
* Use DEL for unmapping grefs instead of PUT
* Rname from xen_netif_gref_alloc to xen_netif_gref
* Add 'status&#
oking
up on our mappings table. Thoughts, or perhaps people may prefer the one
already described in the series?
Cheers,
Joao Martins (1):
public/io/netif.h: add gref mapping control messages
xen/include/public/io/netif.h | 115 ++
1 file changed, 1
[ Realized that I didn't CC the maintainers,
so doing that now, +Linux folks +PV interfaces czar
Sorry for the noise! ]
On 09/08/2017 09:49 AM, Joao Martins wrote:
> [Forgot two important details regarding Xenbus states]
> On 09/07/2017 05:53 PM, Joao Martins wrote:
>> Hey!
&
On 09/14/2017 05:10 PM, Wei Liu wrote:
> On Thu, Sep 07, 2017 at 05:53:54PM +0100, Joao Martins wrote:
>> Hey!
>>
>> We wanted to brought up this small proposal regarding the lack of
>> parameterization on PV devices on Xen.
>>
>> Currently users don&#x
On 09/15/2017 12:34 PM, Juergen Gross wrote:
> On 15/09/17 13:19, Wei Liu wrote:
>> On Thu, Sep 14, 2017 at 05:18:44PM +0100, Joao Martins wrote:
>>> On 09/14/2017 05:10 PM, Wei Liu wrote:
>>>> On Thu, Sep 07, 2017 at 05:53:54PM +0100, Joao Martins wrote:
>&
On Mon, Sep 18, 2017 at 09:45:06AM +, Paul Durrant wrote:
> > -Original Message-
> > From: Joao Martins [mailto:joao.m.mart...@oracle.com]
> > Sent: 13 September 2017 19:11
> > To: Xen-devel
> > Cc: Wei Liu ; Paul Durrant ;
> > Konrad Rzeszutek
On Mon, Sep 18, 2017 at 09:53:18AM +, Paul Durrant wrote:
> > -Original Message-
> > From: Joao Martins [mailto:joao.m.mart...@oracle.com]
> > Sent: 13 September 2017 19:11
> > To: Xen-devel
> > Cc: Wei Liu ; Paul Durrant ;
> > Konrad Rzeszutek
On Mon, Sep 18, 2017 at 12:11:04PM +, Paul Durrant wrote:
> > -Original Message-
> > From: Joao Martins [mailto:joao.m.mart...@oracle.com]
> > Sent: 18 September 2017 12:56
> > To: Paul Durrant
> > Cc: Xen-devel ; Wei Liu ;
> > Konrad Rzeszutek
On 09/18/2017 08:59 PM, Konrad Rzeszutek Wilk wrote:
> On Thu, Sep 14, 2017 at 05:08:18PM +0100, Joao Martins wrote:
>> [ Realized that I didn't CC the maintainers,
>> so doing that now, +Linux folks +PV interfaces czar
>> Sorry for the noise! ]
>>
>> On 0
added are:
- Add grefs to be mapped by backend
- Remove grefs mappings (If they are not in use)
- Get maximum amount of grefs kept mapped.
Signed-off-by: Joao Martins
---
v4:
* Declare xen_netif_gref parameters are input or output.
* Clarify status field and that it doesn't require to b
Hey,
This is v4 taking into consideration all comments received from v3 (changelog
in the first patch). The specification is right after the diffstat.
Reference implementation also here (on top of net-next):
https://github.com/jpemartins/linux.git xen-net-stg-gnts-v3
Cheers,
Joao Martins (1
vcpu_time_info struct.
Signed-off-by: Joao Martins
---
Changes since v1:
* Check flags ahead to see if the primary clock can use
PVCLOCK_TSC_STABLE_BIT even if secondary registration fails.
(Comments from Boris)
* Remove addr, addr variables;
* Change first pr_debug to pr_warn;
* Change last
BI changes.
Changelog since v1 is included in individual patches.
Any comments/suggestions are welcome.
Thanks,
Joao
Joao Martins (3):
x86/pvclock: add setter for pvclock_pvti_cpu0_va
x86/xen/time: setup vcpu 0 time info page
MAINTAINERS: xen, kvm: track pvclock-abi.h changes
M
peers noticed when such changes happen.
Signed-off-by: Joao Martins
Acked-by: Juergen Gross
---
Out of the two options (and provided I was given a choice) I choose the
originally posted because this is so far the only ABI shared between Xen/KVM.
Whenever we have more things shared it would probably de
and moves pvclock_pvti_cpu0_va
to pvclock, which is a more generic place to have it; and would
allow other PV clocksources to use it, such as Xen.
Signed-off-by: Joao Martins
Acked-by: Andy Lutomirski
---
Changes since v1:
* Rebased: the only conflict was that I had move the export
pvclock_pvti_cpu0_va()
On 09/26/2017 10:32 AM, Juergen Gross wrote:
> On 22/09/17 18:25, Joao Martins wrote:
[snip]
>> +static void xen_setup_vsyscall_time_info(void)
>> +{
>> +struct vcpu_register_time_memory_area t;
>> +struct pvclock_vsyscall_time_info *ti;
>> +st
and moves pvclock_pvti_cpu0_va
to pvclock, which is a more generic place to have it; and would
allow other PV clocksources to use it, such as Xen.
Signed-off-by: Joao Martins
Acked-by: Andy Lutomirski
---
Changes since v1:
* Rebased: the only conflict was that I had move the export
pvclock_pvti_cpu0_va()
vcpu_time_info struct.
Signed-off-by: Joao Martins
---
Changes since v2:
(Comments from Juergen)
* Omit the blan after the cast on all 3 occurrences.
* Change last VCLOCK_PVCLOCK message to be more descriptive
* Sync the complete vcpu.h header instead of just adding the
needed one. (IOW adding
hes.
(only patch 2 changed in this version)
Thanks,
Joao
Joao Martins (3):
x86/pvclock: add setter for pvclock_pvti_cpu0_va
x86/xen/time: setup vcpu 0 time info page
MAINTAINERS: xen, kvm: track pvclock-abi.h changes
MAINTAINERS| 2 +
arch/x86/include/asm/pvclo
peers noticed when such changes happen.
Signed-off-by: Joao Martins
Acked-by: Juergen Gross
---
In the end, I choose the originally posted because this is so far the
only ABI shared between Xen/KVM. Therefore whenever we have more things
shared it would deserve its own place in MAINTAINERS file. I
peers noticed when such changes happen.
Signed-off-by: Joao Martins
Acked-by: Juergen Gross
---
In the end, I choose the originally posted because this is so far the
only ABI shared between Xen/KVM. Therefore whenever we have more things
shared it would deserve its own place in MAINTAINERS file. I
1 - 100 of 353 matches
Mail list logo
