core or socket only.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit.c | 2 ++
xen/common/sched_credit2.c | 7 +++
xen/common/sched_null.c| 3 +++
xen/common/sched_rt.c | 2 ++
xen/common/schedule.c | 15 +++
xen/include/xen/sched-if.h | 8
xen
In preparation of core scheduling let the percpu pointer
schedule_data.curr point to a strct sched_item instead of the related
vcpu. At the same time rename the per-vcpu scheduler specific structs
to per-item ones.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 2 +-
xen
ctive previously, so save the previous item pointer
in the per-schedule resource area and use its value being non-NULL as
a hint whether context_saved() should be called.
Signed-off-by: Juergen Gross
---
RFC V2: new patch (Andrew Cooper)
---
xen/arch/x86/domain.c | 21 ++
xen/c
Switch credit scheduler completely from vcpu to sched_item usage.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit.c | 504 +++---
1 file changed, 251 insertions(+), 253 deletions(-)
diff --git a/xen/common/sched_credit.c b/xen/common
Switch credit2 scheduler completely from vcpu to sched_item usage.
As we are touching lots of lines remove some white space at the end of
the line, too.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit2.c | 820 ++---
1 file changed, 403
This prepares support of larger scheduling granularities, e.g. core
scheduling.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 6 ++---
xen/common/sched_credit.c | 14 +--
xen/common/sched_credit2.c| 24 +--
xen/common/sched_null.c | 8
On 06/05/2019 08:55, Juergen Gross wrote:
> Add support for core- and socket-scheduling in the Xen hypervisor.
The series is available under:
github.com/jgross1/xen.git sched-rfc-v2
Juergen
___
Xen-devel mailing list
Xen-devel@lists.xenproject.
On 05/05/2019 15:27, Marek Marczykowski-Górecki wrote:
> Hi,
>
> I have a machine that allocate vesa LFB above 4GB, as reported by UEFI
> GOP. At 0x40 to be specific.
> vga_console_info.u.vesa_lfb.lfb_base is a 32bit field, so it gets
> truncated, leading to all kind of memory corruptions
On 03/05/2019 17:04, Roger Pau Monne wrote:
> There's no reason to request physically contiguous memory for those
> allocations.
>
> Reported-by: Ian Jackson
> Signed-off-by: Roger Pau Monné
Reviewed-by: Juergen Gross
Juergen
_
On 06/05/2019 10:27, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> @@ -207,6 +250,141 @@ static inline void sched_free_domdata(const struct
>> scheduler *s,
>> ASSERT(!data);
>> }
>>
>> +static inline void *sched_alloc_pdata(const struct scheduler *s, int cpu)
>> +{
>> +
On 06/05/2019 10:37, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> --- a/xen/common/schedule.c
>> +++ b/xen/common/schedule.c
>> @@ -808,21 +808,24 @@ static void vcpu_migrate_finish(struct vcpu *v)
>> }
>>
>> /*
>> - * Force a VCPU through a deschedule/reschedule path.
>> - * For ex
On 06/05/2019 10:57, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> --- a/xen/arch/x86/setup.c
>> +++ b/xen/arch/x86/setup.c
>> @@ -1701,6 +1701,8 @@ void __init noreturn __start_xen(unsigned long mbi_p)
>> printk(XENLOG_INFO "Parked %u CPUs\n", num_parked);
>> smp_cpus_done
On 06/05/2019 12:01, Jan Beulich wrote:
On 06.05.19 at 11:23, wrote:
>> On 06/05/2019 10:57, Jan Beulich wrote:
>> On 06.05.19 at 08:56, wrote:
void scheduler_percpu_init(unsigned int cpu)
{
struct scheduler *sched = per_cpu(scheduler, cpu);
struct sched_r
On 06/05/2019 13:58, Jan Beulich wrote:
On 06.05.19 at 12:20, wrote:
>> On 06/05/2019 12:01, Jan Beulich wrote:
>> On 06.05.19 at 11:23, wrote:
On 06/05/2019 10:57, Jan Beulich wrote:
> . Yet then I'm a little puzzled by its use here in the first place.
> Generally I think f
On 06/05/2019 15:14, Jan Beulich wrote:
On 06.05.19 at 14:23, wrote:
>> On 06/05/2019 13:58, Jan Beulich wrote:
>> On 06.05.19 at 12:20, wrote:
On 06/05/2019 12:01, Jan Beulich wrote:
On 06.05.19 at 11:23, wrote:
>> On 06/05/2019 10:57, Jan Beulich wrote:
>>> . Yet
On 06/05/2019 16:50, Marek Marczykowski-Górecki wrote:
> On some machines (for example Thinkpad P52), UEFI GOP reports
> framebuffer located above 4GB (0x40 on that machine). This
> address does not fit in {xen,dom0}_vga_console_info.u.vesa_lfb.lfb_base
> field, which is 32bit. The overflow
On 06/05/2019 17:32, Marek Marczykowski-Górecki wrote:
> On Mon, May 06, 2019 at 05:15:19PM +0200, Juergen Gross wrote:
>> On 06/05/2019 16:50, Marek Marczykowski-Górecki wrote:
>>> diff --git a/xen/include/public/xen.h b/xen/include/public/xen.h
>>> index ccdffc0..b
On 07/05/2019 15:36, Wei Liu wrote:
> It broke Windows installation. More investigation is needed. Don't
> enable it by default.
>
> Signed-off-by: Wei Liu
Thanks!
Juergen
___
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenpr
data of the old
scheduler.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit2.c | 23 +++
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c
index 6958b265fc..9c1c3b4e08 100644
--- a/xen/common/sched_cred
On 08/05/2019 13:39, George Dunlap wrote:
> The "nesting" section in the MAINTAINERS file was not initially
> intended to describe the check-in policy for patches, but only how
> nesting worked; but since there was no check-in policy, it has been
> acting as a de-facto policy.
>
> One problem with
On 06/05/2019 12:01, Jan Beulich wrote:
On 06.05.19 at 11:23, wrote:
>> And that was mentioned in the cover letter: cpu hotplug is not yet
>> handled (hence the RFC status of the series).
>>
>> When cpu hotplug is being added it might be appropriate to switch the
>> scheme as you suggested. R
On 08/05/2019 18:24, George Dunlap wrote:
> On 5/6/19 7:56 AM, Juergen Gross wrote:
>> Instead of using the SCHED_OP() macro to call the different scheduler
>> specific functions add inline wrappers for that purpose.
>>
>> Signed-off-by: Juergen Gross
>
> This se
On 08/05/2019 18:35, George Dunlap wrote:
> On 5/6/19 7:56 AM, Juergen Gross wrote:
>> In order to prepare core- and socket-scheduling use a new struct
>> sched_item instead of struct vcpu for interfaces of the different
>> schedulers.
>>
>> Rename the per-sc
On 09/05/2019 07:56, Dario Faggioli wrote:
> On Thu, 2019-05-09 at 07:36 +0200, Juergen Gross wrote:
>> On 08/05/2019 18:35, George Dunlap wrote:
>>> On 5/6/19 7:56 AM, Juergen Gross wrote:
>>>>
>>> 'item' really seems like the wrong word here. To
On 09/05/2019 12:04, George Dunlap wrote:
> On 5/9/19 6:32 AM, Juergen Gross wrote:
>> On 08/05/2019 18:24, George Dunlap wrote:
>>> On 5/6/19 7:56 AM, Juergen Gross wrote:
>>>> Instead of using the SCHED_OP() macro to call the different scheduler
>>>>
On 09/05/2019 13:50, Jan Beulich wrote:
>>>> On 09.05.19 at 12:56, wrote:
>> On 09/05/2019 12:04, George Dunlap wrote:
>>> On 5/9/19 6:32 AM, Juergen Gross wrote:
>>>> On 08/05/2019 18:24, George Dunlap wrote:
>>>>> On 5/6/19 7:56 AM, Jue
On 09/05/2019 14:31, Jan Beulich wrote:
>>>> On 09.05.19 at 14:03, wrote:
>> On 09/05/2019 13:50, Jan Beulich wrote:
>>>>>> On 09.05.19 at 12:56, wrote:
>>>> On 09/05/2019 12:04, George Dunlap wrote:
>>>>> On 5/9/19 6:32 AM, Juerg
nsole.dom0.info_size), Linux can detect when
> this field is present and use it appropriately then.
>
> Signed-off-by: Marek Marczykowski-Górecki
Public interface part:
Reviewed-by: Juergen Gross
Juergen
___
Xen-devel mailing list
Xen-devel@list
On 08/05/2019 12:28, Ross Lagerwall wrote:
> During a suspend/resume, the xenwatch thread waits for all outstanding
> xenstore requests and transactions to complete. This does not work
> correctly for transactions started by userspace because it waits for
> them to complete after freezing userspace
On 10/05/2019 10:53, Jan Beulich wrote:
On 08.05.19 at 16:36, wrote:
>> On 06/05/2019 12:01, Jan Beulich wrote:
>> On 06.05.19 at 11:23, wrote:
And that was mentioned in the cover letter: cpu hotplug is not yet
handled (hence the RFC status of the series).
When cpu ho
On 10/05/2019 12:29, Dario Faggioli wrote:
> On Fri, 2019-05-10 at 11:00 +0200, Juergen Gross wrote:
>> On 10/05/2019 10:53, Jan Beulich wrote:
>>>>>> On 08.05.19 at 16:36, wrote:
>>>>
>>>> With sched-gran=core or sched-gran=socket offlining a
On 11/05/2019 13:20, osstest service owner wrote:
> flight 135931 xen-unstable real [real]
> http://logs.test-lab.xenproject.org/osstest/logs/135931/
>
> Regressions :-(
>
> Tests which did not succeed and are blocking,
> including tests which could not be run:
> test-amd64-i386-xl 23
On 11/05/2019 19:57, Wei Liu wrote:
> On Sat, May 11, 2019 at 02:29:53PM +0200, Juergen Gross wrote:
>> On 11/05/2019 13:20, osstest service owner wrote:
>>> flight 135931 xen-unstable real [real]
>>> http://logs.test-lab.xenproject.org/osstest/logs/135931/
>>&g
Committers,
The commit moratorium is lifted.
Juergen
___
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
On 13/05/2019 17:34, Wei Liu wrote:
> Hello
>
> Seeing that you were the last people who changed blktap2 in a meaningful
> way: do you use it at all?
Not me. I was only changing it to comply with the rest of the build
(adding pkg-config file).
I SUSE builds (SLE, openSUSE) it is not configured.
Linus,
Please git pull the following tag:
git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip.git
for-linus-5.2b-rc1-tag
xen: fixes and features for 5.2-rc1
It contains:
- some minor cleanups
- 2 small corrections for Xen on ARM
- 2 fixes for Xen PVH guest support
- a patch for a new comman
The patch "libxl: add helper function to set device_model_version"
breaks creating any domain for me.
Creating a PV or HVM guest will trigger the assertion
assert(b_info->device_model_version);
added by that patch. Removing this assert() will then trigger
xl: libxl.c:339: libxl_defbool_val: Ass
On 16/05/2019 10:27, Olaf Hering wrote:
> Am Thu, 16 May 2019 10:09:38 +0200
> schrieb Juergen Gross :
>
>> Has this patch ever been tested to work?
>
> With PV only. I will have a look.
I can't even start a PV domU (maybe that is due t
On 16/05/2019 10:30, Olaf Hering wrote:
> Am Thu, 16 May 2019 10:27:06 +0200
> schrieb Olaf Hering :
>
>> Am Thu, 16 May 2019 10:09:38 +0200
>> schrieb Juergen Gross :
>>> Has this patch ever been tested to work?
>> With PV only. I will have a
On 16/05/2019 10:40, Olaf Hering wrote:
> Am Thu, 16 May 2019 10:09:38 +0200
> schrieb Juergen Gross :
>
>> assert(b_info->device_model_version);
>
> Is the codepath perhaps coming from libxl_domain_need_memory?
Yes:
(gdb) bt
#0 0x76f1b0c7 in raise ()
On 16/05/2019 11:11, Wei Liu wrote:
> Two lines were erroneously deleted during rebase which caused domain
> destruction to fail.
>
> Signed-off-by: Wei Liu
Tested-by: Juergen Gross
Reviewed-by: Juergen Gross
Juergen
___
Xen-devel
On 16/05/2019 14:20, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> --- a/xen/common/schedule.c
>> +++ b/xen/common/schedule.c
>> @@ -314,14 +314,42 @@ static struct sched_item *sched_alloc_item(struct vcpu
>> *v)
>> return NULL;
>> }
>>
>> -int sched_init_vcpu(struct vcpu *v, un
On 16/05/2019 14:30, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> --- a/xen/arch/x86/domain.c
>> +++ b/xen/arch/x86/domain.c
>> @@ -1619,6 +1619,37 @@ static inline bool need_full_gdt(const struct domain
>> *d)
>> return is_pv_domain(d) && !is_idle_domain(d);
>> }
>>
>> +static
differ twice.
>>
>> Cache the need_full_gdt(nd) value in a local variable.
>>
>> Signed-off-by: Juergen Gross
>
> To be honest I'm not entirely convinced this is a good move. But
> since you've done the work, and since the larger source size is
> h
On 16/05/2019 15:05, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> --- a/xen/arch/x86/domain.c
>> +++ b/xen/arch/x86/domain.c
>> @@ -154,6 +154,24 @@ static void idle_loop(void)
>> }
>> }
>>
>> +/*
>> + * Idle loop for siblings of active schedule items.
>> + * We don't do any sta
On 16/05/2019 16:41, Jan Beulich wrote:
On 16.05.19 at 15:51, wrote:
>> On 16/05/2019 15:05, Jan Beulich wrote:
>> On 06.05.19 at 08:56, wrote:
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
@@ -154,6 +154,24 @@ static void idle_loop(void)
}
}
On 17/05/2019 08:57, Jan Beulich wrote:
On 17.05.19 at 07:13, wrote:
>> On 16/05/2019 16:41, Jan Beulich wrote:
>> On 16.05.19 at 15:51, wrote:
On 16/05/2019 15:05, Jan Beulich wrote:
On 06.05.19 at 08:56, wrote:
>> --- a/xen/arch/x86/domain.c
>> +++ b/xen/arch/x86
On 17/05/2019 10:22, Jan Beulich wrote:
On 17.05.19 at 09:48, wrote:
>> On 17/05/2019 08:57, Jan Beulich wrote:
>> On 17.05.19 at 07:13, wrote:
On 16/05/2019 16:41, Jan Beulich wrote:
On 16.05.19 at 15:51, wrote:
>> As with core scheduling we can be sure the other thre
Avoid that by removing the cpu from the list of active cpus in credit2
>> data first.
>>
>> The opposite problem is occurring when removing a cpu from a cpupool:
>> init_pdata() of credit2 will access the per-cpu data of the old
>> scheduler.
>>
>> Signed-off-
of the just moved cpu. This
>>>> will (depending on the new scheduler) either clobber the data of the
>>>> new scheduler or in case of sched_rt lead to a crash.
>>>>
>>>> Avoid that by removing the cpu from the list of active cpus in credit2
>>>&
On 20/05/2019 12:46, Ian Jackson wrote:
> This has been true for some time. osstest has been sending automatic
> mails including bisection reports like the one below, for about a
> month.
>
> osstest's "tested and good" branch of linux-3.18 is 190 days behind
> the upstream tip.
>
> Do we care a
There is no user of the preempt_count for non-debug builds. This series
removes it for that case.
Juergen Gross (3):
xen: drop in_atomic()
xen: drop preempt_count() for non-debug builds
xen: build common/preempt.c only with CONFIG_DEBUG
xen/arch/x86/hvm/hvm.c| 16
xen
.
Signed-off-by: Juergen Gross
---
xen/arch/x86/hvm/hvm.c| 16
xen/common/preempt.c | 5 -
xen/include/xen/preempt.h | 2 --
3 files changed, 23 deletions(-)
diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
index 8993c2aa57..7b36765b97 100644
--- a/xen
preempt_count() and the associated per-cpu variable __preempt_count
are tested in debug build only. So drop them for non-debug builds.
Signed-off-by: Juergen Gross
---
xen/common/preempt.c | 2 +-
xen/include/xen/preempt.h | 6 +-
2 files changed, 6 insertions(+), 2 deletions(-)
diff
There is nothing left in common/preempt.c in non-debug builds.
Signed-off-by: Juergen Gross
---
xen/common/Makefile | 2 +-
xen/common/preempt.c | 2 --
2 files changed, 1 insertion(+), 3 deletions(-)
diff --git a/xen/common/Makefile b/xen/common/Makefile
index 33d03b862f..d6576a3fc3 100644
On 22/05/2019 12:00, Andrew Cooper wrote:
> On 22/05/2019 10:45, Juergen Gross wrote:
>> preempt_count() and the associated per-cpu variable __preempt_count
>> are tested in debug build only. So drop them for non-debug builds.
>>
>> Signed-off-by: Juergen Gross
>
&g
On 22/05/2019 12:12, Jan Beulich wrote:
On 22.05.19 at 11:45, wrote:
>> @@ -26,9 +28,11 @@ DECLARE_PER_CPU(unsigned int, __preempt_count);
>> preempt_count()--; \
>> } while (0)
>>
>> -#ifndef NDEBUG
>> void ASSERT_NOT_IN_ATOMIC(void);
>> +
>> #else
>> +#def
On 22/05/2019 12:10, Jan Beulich wrote:
On 22.05.19 at 11:45, wrote:
>> --- a/xen/arch/x86/hvm/hvm.c
>> +++ b/xen/arch/x86/hvm/hvm.c
>> @@ -3185,22 +3185,6 @@ static enum hvm_translation_result __hvm_copy(
>>
>> ASSERT(is_hvm_vcpu(v));
>>
>> -/*
>> - * XXX Disable for 4.1.0:
On 22/05/2019 12:18, Jan Beulich wrote:
>>>> On 22.05.19 at 12:00, wrote:
>> On 22/05/2019 10:45, Juergen Gross wrote:
>>> preempt_count() and the associated per-cpu variable __preempt_count
>>> are tested in debug build only. So drop them for non-debug bu
On 22/05/2019 14:34, Jan Beulich wrote:
On 22.05.19 at 12:19, wrote:
>> On 22/05/2019 12:10, Jan Beulich wrote:
>> On 22.05.19 at 11:45, wrote:
--- a/xen/arch/x86/hvm/hvm.c
+++ b/xen/arch/x86/hvm/hvm.c
@@ -3185,22 +3185,6 @@ static enum hvm_translation_result __hvm_copy(
>
On 22/05/2019 12:10, Jan Beulich wrote:
On 22.05.19 at 11:45, wrote:
>> --- a/xen/arch/x86/hvm/hvm.c
>> +++ b/xen/arch/x86/hvm/hvm.c
>> @@ -3185,22 +3185,6 @@ static enum hvm_translation_result __hvm_copy(
>>
>> ASSERT(is_hvm_vcpu(v));
>>
>> -/*
>> - * XXX Disable for 4.1.0:
On 24/05/2019 08:38, Jan Beulich wrote:
On 24.05.19 at 07:41, wrote:
>> On 22/05/2019 12:10, Jan Beulich wrote:
>> On 22.05.19 at 11:45, wrote:
--- a/xen/arch/x86/hvm/hvm.c
+++ b/xen/arch/x86/hvm/hvm.c
@@ -3185,22 +3185,6 @@ static enum hvm_translation_result __hvm_copy(
>
vm_event_claim_slot() returns -EOPNOTSUPP for an uninitialized ring
since commit 15e4dd5e866b43bbc ("common/vm_event: Initialize vm_event
lists on domain creation"), but the callers test for -ENOSYS.
Correct the callers.
Signed-off-by: Juergen Gross
---
xen/arch/x86/mm/p2m.c | 2 +-
On 24/05/2019 16:05, Razvan Cojocaru wrote:
> On 5/24/19 4:15 PM, Juergen Gross wrote:
>> vm_event_claim_slot() returns -EOPNOTSUPP for an uninitialized ring
>> since commit 15e4dd5e866b43bbc ("common/vm_event: Initialize vm_event
>> lists on domain creation"), b
On 24/05/2019 16:11, Jan Beulich wrote:
On 24.05.19 at 15:15, wrote:
>> --- a/xen/arch/x86/mm/p2m.c
>> +++ b/xen/arch/x86/mm/p2m.c
>> @@ -1705,7 +1705,7 @@ void p2m_mem_paging_populate(struct domain *d,
>> unsigned long gfn_l)
>>
>> /* We're paging. There should be a ring */
>> i
vm_event_claim_slot() returns -EOPNOTSUPP for an uninitialized ring
since commit 15e4dd5e866b43bbc ("common/vm_event: Initialize vm_event
lists on domain creation"), but the callers test for -ENOSYS.
Correct the callers.
Signed-off-by: Juergen Gross
---
V2: add blank line (J
hen Hemminger
> Cc: Sasha Levin
> Cc: Thomas Gleixner
> Cc: Ingo Molnar
> Cc: Borislav Petkov
> Cc: x...@kernel.org
> Cc: Juergen Gross
> Cc: Paolo Bonzini
> Cc: Dave Hansen
> Cc: Andy Lutomirski
> Cc: Peter Zijlstra
> Cc: Boris Ostrovsky
> Cc:
drivers/xen/pvcalls-front.c: In function pvcalls_front_recvmsg:
> drivers/xen/pvcalls-front.c:638:25: warning: variable bedata set but not used
> [-Wunused-but-set-variable]
>
> They are never used since introduction.
>
> Signed-off-by: YueH
scarded, return OK
> since no changes were made anyway.
>
> This only affects users of the xenbus file interface. In-kernel users of
> xenbus are assumed to be well-behaved and complete all transactions
> before freezing.
>
> Signed-off-by: Ross Lagerwall
Reviewed-by: Juergen G
The Xen tmem (transcendent memory) driver can be removed, as the
related Xen hypervisor feature never made it past the "experimental"
state and will be removed in future Xen versions (>= 4.13).
The xen-selfballoon driver depends on tmem, so it can be removed, too.
Signed-off-by:
Tmem has been an experimental Xen feature which has been dropped
recently due to security problems and lack of maintainership.
So it is time now to drop it in Linux kernel, too.
Juergen Gross (3):
xen: remove tmem driver
mm: remove cleancache.c
mm: remove tmem specifics from frontswap
:
- ARM is building now
- HVM domains are working now
- idling will always be done with idle_vcpu active
- other small changes see individual patches
Juergen Gross (60):
xen/sched: only allow schedulers with all mandatory functions
available
xen/sched: add inline wrappers for calling per-sc
new struct scheduler and struct schedule_data
values in the percpu area in schedule_cpu_switch() instead of the
schedulers, too.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 14 ++
xen/common/sched_credit.c | 13 ++---
xen/common/sched_credit2.c | 15
Add an identifier to sched_unit. For now it will be the same as the
related vcpu_id.
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 3 ++-
xen/include/xen/sched.h | 1 +
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/xen/common/schedule.c b/xen/common/schedule.c
index
This prepares support of larger scheduling granularities, e.g. core
scheduling.
While at it move sched_has_urgent_vcpu() from include/asm-x86/cpuidle.h
into schedule.c removing the need for including sched-if.h in
cpuidle.h and multiple other C sources.
Signed-off-by: Juergen Gross
---
V1: move
Allocate a struct sched_unit for each vcpu. This removes the need to
have it locally on the stack in schedule.c.
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 67 +++--
xen/include/xen/sched.h | 2 ++
2 files changed, 33 insertions
Instead of returning a physical cpu number let pick_cpu() return a
scheduler resource instead. Rename pick_cpu() to pick_resource() to
reflect that change.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 12 ++--
xen/common/sched_credit.c| 16
xen
Add a pointer to the domain to struct sched_unit in order to avoid
having to dereference the vcpu pointer of struct sched_unit to find
the related domain.
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 3 ++-
xen/include/xen/sched.h | 1 +
2 files changed, 3 insertions(+), 1
for_each_sched_unit_vcpu() which
will iterate over all vcpus if a sched_unit (right now only one). This
will be needed later for larger scheduling granularity (e.g. cores).
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 56 ++---
xen/include
Rename vcpu_schedule_[un]lock[_irq]() to unit_schedule_[un]lock[_irq]()
and let it take a sched_unit pointer instead of a vcpu pointer as
parameter.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit.c | 17
xen/common/sched_credit2.c | 40
switched to schedule unit, too.
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 102 +-
1 file changed, 59 insertions(+), 43 deletions(-)
diff --git a/xen/common/schedule.c b/xen/common/schedule.c
index 12f9852786..8121da15c6 100644
--- a/xen
Some functions of struct scheduler are mandatory. Test those in the
scheduler initialization loop to be present and drop schedulers not
complying.
Signed-off-by: Juergen Gross
---
V1: new patch
---
xen/common/schedule.c | 26 +-
1 file changed, 25 insertions(+), 1
Instead of dynamically decide whether the previous vcpu was using full
or default GDT just add a percpu variable for that purpose. This at
once removes the need for testing vcpu_ids to differ twice.
Cache the need_full_gdt(nd) value in a local variable.
Signed-off-by: Juergen Gross
Reviewed-by
Prepare supporting multiple cpus per scheduling resource by allocating
the cpumask per resource dynamically.
Modify sched_res_mask to have only one bit per scheduling resource set.
Signed-off-by: Juergen Gross
---
V1: new patch (carved out from other patch)
---
xen/common/schedule.c | 16
mains will
have initially all nodes set, so this is no different behavior compared
to today.
To be able to use const struct domain * make cpupool_domain_cpumask()
take a const domain pointer, too.
Signed-off-by: Juergen Gross
Acked-by: Andrew Cooper
---
RFC V2: add ASSERT(), modify error messa
core or socket only.
Signed-off-by: Juergen Gross
---
V1: add accessor functions
---
xen/common/sched_credit.c | 2 ++
xen/common/sched_credit2.c | 7 +++
xen/common/sched_null.c| 3 +++
xen/common/sched_rt.c | 2 ++
xen/common/schedule.c | 15 +++
xen/include/xen
}_pause_flags[_atomic]() to modify pause_flags of the
associated vcpu(s)
- sched_idle_unit() to get the sched_unit pointer of the idle vcpu of a
specific physical cpu
Signed-off-by: Juergen Gross
---
xen/common/sched_credit.c | 3 +--
xen/common/schedule.c | 19
xen/include
Use sched_units instead of vcpus in schedule(). This includes the
introduction of sched_unit_runstate_change() as a replacement of
vcpu_runstate_change() in schedule().
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 70 +--
1 file
Switch rt scheduler completely from vcpu to sched_unit usage.
Signed-off-by: Juergen Gross
---
xen/common/sched_rt.c | 356 --
1 file changed, 174 insertions(+), 182 deletions(-)
diff --git a/xen/common/sched_rt.c b/xen/common/sched_rt.c
index
Switch arinc653 scheduler completely from vcpu to sched_unit usage.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 208 +---
1 file changed, 101 insertions(+), 107 deletions(-)
diff --git a/xen/common/sched_arinc653.c b/xen/common
Add counters to struct sched_unit summing up runstates of associated
vcpus.
Signed-off-by: Juergen Gross
---
RFC V2: add counters for each possible runstate
---
xen/common/schedule.c | 5 +
xen/include/xen/sched.h | 2 ++
2 files changed, 7 insertions(+)
diff --git a/xen/common
Let the schedulers put a sched_unit pointer into struct task_slice
instead of a vcpu pointer.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 8
xen/common/sched_credit.c | 4 ++--
xen/common/sched_credit2.c | 4 ++--
xen/common/sched_null.c | 12
sched_move_irqs() should work on a sched_unit as that is the unit
moved between cpus.
Rename the current function to vcpu_move_irqs() as it is still needed
in schedule().
Signed-off-by: Juergen Gross
---
xen/common/schedule.c | 18 +-
1 file changed, 13 insertions(+), 5
pdated whenever the is_running indicator is changed.
Use that new field in the schedulers instead of the similar vcpu field.
Signed-off-by: Juergen Gross
---
RFC V2: fix arm build, don't drop v->is_running
---
xen/common/sched_credit.c | 12 +++-
xen/common/sched_
This prepares making the different schedulers vcpu agnostic.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 4 ++--
xen/common/sched_credit.c | 6 +++---
xen/common/sched_credit2.c | 10 +-
xen/common/sched_null.c | 4 ++--
xen/common/sched_rt.c | 4
.
Signed-off-by: Juergen Gross
---
xen/common/sched_arinc653.c | 20 +++-
xen/common/sched_credit.c | 25 +++--
xen/common/sched_credit2.c | 21 +
xen/common/sched_null.c | 26 --
xen/common/sched_rt.c | 22
In several places there is support for multiple vcpus per sched unit
missing. Add that missing support (with the exception of initial
allocation) and missing helpers for that.
Signed-off-by: Juergen Gross
---
RFC V2: fix vcpu_runstate_helper()
V1: add special handling for idle unit in
Rename the scheduler related perf counters from vcpu* to unit* where
appropriate.
Signed-off-by: Juergen Gross
---
xen/common/sched_credit.c| 32
xen/common/sched_credit2.c | 18 +-
xen/common/sched_null.c | 18 +-
xen
Where appropriate switch from for_each_vcpu() to for_each_sched_unit()
in order to prepare core scheduling.
Signed-off-by: Juergen Gross
---
xen/common/domain.c | 9 ++---
xen/common/schedule.c | 109 ++
2 files changed, 60 insertions(+), 58
Instead of using the SCHED_OP() macro to call the different scheduler
specific functions add inline wrappers for that purpose.
Signed-off-by: Juergen Gross
---
RFC V2: new patch (Andrew Cooper)
V1: use conditional operator (Jan Beulich, Dario Faggioli)
drop no longer needed ASSERT()s
401 - 500 of 7454 matches
Mail list logo
