Re: [Xen-devel] [RFC PATCH v4] xen: credit2: provide custom option to create runqueue

2017-12-07 Thread Dario Faggioli
On Tue, 2017-12-05 at 21:58 +0530, Praveen Kumar wrote: > On Tue, Sep 5, 2017 at 11:26 PM, Dario Faggioli > <dario.faggi...@citrix.com> wrote: > > > > I'm asking because I do have it half done myself, and it would not > > take > > too much time to me to

Re: [Xen-devel] [PATCH v6 00/16] xen: common: rbtree: ported updates from Linux tree

2017-12-07 Thread Dario Faggioli
. BTW, can you update my email address to raist...@linux.it ? Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli signature.asc Description: This is a digitall

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-10 Thread Dario Faggioli
instead than to CPU_STARTING, SCHED_OP(init_pdata) wouldn't be called, without having to fiddle with priorities. Is there any reason why you can't do it that way? It would look more natural to me, and it's definitely going to be easier debug and maintain (e.g., look at how many callbacks CPU_UP_PRE

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-10 Thread Dario Faggioli
details about that, and we'll try to figure out what's causing it. Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description: This is

Re: [Xen-devel] [PATCH v3 06/10] xen/common: Restore IRQ affinity when hotplugging a pCPU

2018-05-10 Thread Dario Faggioli
. > This patch adds the restoration of IRQ affinity when a pCPU is > hotplugged. > > Signed-off-by: Mirela Simonovic <mirela.simono...@aggios.com> > > --- > CC: George Dunlap <george.dun...@eu.citrix.com> > CC: Dario Faggioli <dfaggi...@suse.com> > R

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-10 Thread Dario Faggioli
CPU_STARTING xx> CPU_DIDNT_START --> CPU_UP_CANCEL At which point, e.g., from the scheduler point of view, we can try to put a call to SCHED_OP(deinit_pdata) in CPU_DIDNT_START, and that would avoid the problem Mirela is facing, without having to play with priorities. Regards, Dario -- &l

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-10 Thread Dario Faggioli
of that discussion here. > As just saind in the other email, I don't think this is all it's necessary to enable CPU_STARTING to fail. Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli So

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-10 Thread Dario Faggioli
f that is the case, giving different priorities to the notifier, is not a solution. Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ s

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-11 Thread Dario Faggioli
CPU bringup process accordingly. Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.a

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-11 Thread Dario Faggioli
ringup code, which will happen in a leter series." that would also work, I guess. :-) Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-11 Thread Dario Faggioli
On Fri, 2018-05-11 at 12:41 +0200, Mirela Simonovic wrote: > Hi Dario, > Hi, > On Thu, May 10, 2018 at 6:24 PM, Dario Faggioli <dfaggi...@suse.com> > wrote: > > I may very well be missing or misunderstanding something, but I > > continue to think that the problem h

Re: [Xen-devel] [PATCH v3 10/10] xen/arm: Enable errata for secondary CPU on hotplug after the boot

2018-05-11 Thread Dario Faggioli
e. What do you > think? > Well, all I can say is that you should stay away from notifiers priority --especially of the one of cpu_schedule_callback(). As long as you do that, I won't be on your way. :-D Regards, Dario -- <> (Raistlin Majere) -

Re: [Xen-devel] [RFC][PATCH] KVM: APPLES can improve the performance of applications and virtualized systems by up to 49%

2018-05-17 Thread Dario Faggioli
-- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description: This is a digitally signed message part

Re: [Xen-devel] Test for osstest, features used in Qubes OS

2018-05-21 Thread Dario Faggioli
ble > it is? > > That way we get suspend testing 95% effective as quickly as possible, > and we can complete it as we have time. > That sounds a very good plan to me, FWIW. Regards, Dario -- <> (Raistlin Majere)

Re: [Xen-devel] Test for osstest, features used in Qubes OS

2018-05-21 Thread Dario Faggioli
) (although, I did not actually check the boxes in the MA colo, they were just servers from Citrix's lab). There's a (non-perfect) workaround, though, as George suggests, which would allow us to run a "quasi-suspend" test at every flight on every hardware. Regards, Dario -- <> (Rais

Re: [Xen-devel] [PATCH] xen: Add fake_s3 command-line option for testing

2018-05-22 Thread Dario Faggioli
liser, and you need to > > patch > > xen-command-line.markdown. > > Hmm, actually -- would it make more sense to add this as an option to > acpi_sleep? e.g., "acpi_sleep=s3_fake"? (Along with s3_bios and > s3_mode?) > This actually seems a good idea to

Re: [Xen-devel] Test for osstest, features used in Qubes OS

2018-05-22 Thread Dario Faggioli
d the fake suspend test to every > flight. > > Do we want or need to do that test with a guest running ? > Doing it with a guest running would be more complete, I think. I think the best would be to do both, i.e.: - suspend without any guest - (when resumed) start a guest - suspend with a gues

Re: [Xen-devel] [PATCH] xen/credit2: Drop unnecessary bit test

2018-01-11 Thread Dario Faggioli
On Thu, 2018-01-11 at 16:50 +, George Dunlap wrote: > On 01/11/2018 04:48 PM, Andrew Cooper wrote: > > Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com> > > --- > > CC: George Dunlap <george.dun...@eu.citrix.com> > > CC: Dario Faggioli <dfag

[Xen-devel] [PATCH] MAINTAINERS: update my entries to new email address.

2018-01-10 Thread Dario Faggioli
Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: Andrew Cooper <andrew.coop...@citrix.com> Cc: George Dunlap <george.dun...@eu.citrix.com> Cc: Ian Jackson <ian.jack...@eu.citrix.com> Cc: Jan Beulich <jbeul...@suse.com> Cc: Konrad Rzeszutek Wilk <ko

Re: [Xen-devel] sidecar (hvm shim) creation script

2018-01-10 Thread Dario Faggioli
0 gsb: gss: 0002 (XEN) [ 314.149830] ds: es: fs: gs: ss: cs: e008 Also, find attached the various log files. So... what am I doing wrong? :-) Regards, Dario -- <> (Raistlin Majere) -----

Re: [Xen-devel] sidecar (hvm shim) creation script

2018-01-10 Thread Dario Faggioli
fsb: gsb: gss: 0002 (XEN) [ 314.149830] ds: es: fs: gs: ss: cs: e008 Also, find attached the various log files. So... what am I doing wrong? :-) Regards, Dario -- <> (Raistlin Majere) --

Re: [Xen-devel] [PATCH RFC v2 00/12] xen/x86: use per-vcpu stacks for 64 bit pv domains

2018-01-23 Thread Dario Faggioli
n's band-aid applied, but disabled); - - "staging-xpti-on" this is current staging, booted with xpti=true (so, with Jan's band-aid applied, but enabled); - - "4.11 Juergen xpti" this is Juergen's GitHub branch, booted with xpti=true. I'll post more as soon as I will have

Re: [Xen-devel] [PATCH] xen: Credit2: enable fully custom runqueue arrangement

2018-01-24 Thread Dario Faggioli
On Wed, 2018-01-24 at 11:44 +, George Dunlap wrote: > On 09/13/2017 05:21 PM, Dario Faggioli wrote: > > diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen- > > command-line.markdown > > index 9797c8d..dbf5d4c 100644 > > --- a/docs/misc/xen-command-l

Re: [Xen-devel] [PATCH v10 09/11] x86/ctxt: Issue a speculation barrier between vcpu contexts

2018-01-26 Thread Dario Faggioli
; no other load than that caused by the guest). > Ok. > If you want, I could > bring the code I've used for monitoring into patch form and hand > it to you. > If it's not a problem, and whenever you have time, yes, that would be useful, I think. Thanks and Regards, Dario --

Re: [Xen-devel] [PATCH RFC v1 42/74] sched/null: skip vCPUs on the waitqueue that are blocked

2018-01-12 Thread Dario Faggioli
r shim's vCPUs (so, the CPUs of L1 Xen). I'm less fine with rescheduling everyone at every wakeup. Roger, Wei, if/when you want to talk a bit about this, to explain the situation a bit better, so I'll be able to help, feel free to ping me (email or IRC). :-) Regards, Dario -- <> (Raistl

Re: [Xen-devel] [PATCH RFC v1 42/74] sched/null: skip vCPUs on the waitqueue that are blocked

2018-01-12 Thread Dario Faggioli
I think things like this are what we want. As said above for the wakeup case, though, I'd prefer to find a way to avoid that offline vCPUs ends up in the waitqueue, rather than having to skip them. Side note, is_vcpu_online() can be used for the test. Regards, Dario -- <> (Raistlin Majere) ---

Re: [Xen-devel] [PATCH] x86/bitops: Introduce variable/constant pairs for __{set, clear, change}_bit()

2018-01-12 Thread Dario Faggioli
macro, and fix the indentation. > > Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com> > FWIW, Reviewed-by: Dario Faggioli <dfaggi...@suse.com> Regards, Dario -- <> (Raistlin Majere) - Dario Fa

Re: [Xen-devel] [PATCH RFC v1 42/74] sched/null: skip vCPUs on the waitqueue that are blocked

2018-01-12 Thread Dario Faggioli
On Fri, 2018-01-12 at 10:45 +, Roger Pau Monné wrote: > On Fri, Jan 12, 2018 at 10:54:03AM +0100, Dario Faggioli wrote: > > > Err... yes. BTW, either there are a couple of typos in the above > > paragraph, or it's me that can't read it well. Anyway, just to be > > c

Re: [Xen-devel] [PATCH v10 09/11] x86/ctxt: Issue a speculation barrier between vcpu contexts

2018-01-25 Thread Dario Faggioli
On Thu, 2018-01-25 at 16:09 +, Andrew Cooper wrote: > On 25/01/18 15:57, Jan Beulich wrote: > > > > > > > For the record, the overwhelming majority of calls to > > __sync_local_execstate() being responsible for the behavior > > come from invalidate_interrupt(), which suggests to me that > >

Re: [Xen-devel] [PATCH v10 09/11] x86/ctxt: Issue a speculation barrier between vcpu contexts

2018-01-25 Thread Dario Faggioli
On Thu, 2018-01-25 at 19:49 +0100, Dario Faggioli wrote: > On Thu, 2018-01-25 at 16:09 +, Andrew Cooper wrote: > > On 25/01/18 15:57, Jan Beulich wrote: > > > > > > For the record, the overwhelming majority of calls to > > > __sync_local_execstate(

Re: [Xen-devel] [PATCH] xen/sched_rt: Move repl_timer into struct rt_private

2018-01-11 Thread Dario Faggioli
rt_init 219 136 -83 > > As an extra bit of cleanup noticed while making this change, there is > no need > to call cpumask_clear() on an zeroed memory allocation. > > Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com> > Ack

Re: [Xen-devel] [PATCH] xen/sched_rt: Move repl_timer into struct rt_private

2018-01-11 Thread Dario Faggioli
rt_init 219 136 -83 > > As an extra bit of cleanup noticed while making this change, there is > no need > to call cpumask_clear() on an zeroed memory allocation. > > Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com> > Ack

Re: [Xen-devel] [PATCH] xen/credit2: Drop unnecessary bit test

2018-01-12 Thread Dario Faggioli
ing. :-) It's not necessary, as George Acked it already, but feel free to add a: Reviewed-by: Dario Faggioli <dfaggi...@suse.com> Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dar

Re: [Xen-devel] [PATCH v10 09/11] x86/ctxt: Issue a speculation barrier between vcpu contexts

2018-01-26 Thread Dario Faggioli
ance, does not suffer from this issue. In fact, hyperthreading, there, is considered during wakeup/tickling already. Hope this helps clarifying things a bit, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/

Re: [Xen-devel] [PATCH 3/3] xen: Disable ARINC653 scheduler by default for non-DEBUG builds

2018-02-05 Thread Dario Faggioli
ble it by default for > > non-DEBUG builds. > > The more that its prompt says "(EXPERIMENTAL)". > > > Signed-off-by: George Dunlap <george.dun...@citrix.com> > > Acked-by: Jan Beulich <jbeul...@suse.com>

Re: [Xen-devel] [PATCH 2/3] xen: Fix credit1 Kconfig entry

2018-02-05 Thread Dario Faggioli
t; > > Acked-by: Jan Beulich <jbeul...@suse.com> > Reviewed-by: Dario Faggioli <dfaggi...@suse.com> Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggiol

Re: [Xen-devel] [PATCH 1/3] Make credit2 the default scheduler

2018-02-05 Thread Dario Faggioli
ve often said we would switch at the very beginning of a dev cycle, to get as much as osstest and day-by-day testing from developer as possible. Considering we're releasing in June, but freezing in March, do we think it is still early enough? Regards, Dario -- <> (Raistlin Majere)

Re: [Xen-devel] [PATCH 1/2] make xen ocaml safe-strings compliant

2018-02-08 Thread Dario Faggioli
, things work again for me. Regards, Dario PS. There has been a v2 of this patch, I think, but I don't have in my emails any longer, so I'm replying to this one. -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about

Re: [Xen-devel] [PATCH 1/2] make xen ocaml safe-strings compliant

2018-02-09 Thread Dario Faggioli
n some of yours XenRT runs... but really, I'm happy to provide more info. Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature

Re: [Xen-devel] [PATCH 1/3] Make credit2 the default scheduler

2018-02-06 Thread Dario Faggioli
On Tue, 2018-02-06 at 17:02 +, George Dunlap wrote: > On 02/06/2018 06:18 AM, Juergen Gross wrote: > > On 05/02/18 17:53, Dario Faggioli wrote: > > > > > > Considering we're releasing in June, but freezing in March, do we > > > think > > > it i

Re: [Xen-devel] RTDS with extra time issue

2018-02-12 Thread Dario Faggioli
st* *one* real-time task. :-/ Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description: This is a digitally signed message part

Re: [Xen-devel] [PATCH v3 00/17] Alternative Meltdown mitigation

2018-02-12 Thread Dario Faggioli
ne, on Juergen's XPTI series, to mitigate the performance impact a bit. Juergen sent me a patch already, and I'm re-running the benchmarks with that applied. I'll let know how the results ends up looking like. Regards, Dario -- <> (Raistlin Majere) -----

Re: [Xen-devel] RTDS with extra time issue

2018-02-09 Thread Dario Faggioli
ou still see deadline misses? Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description: This is a digitally signed message part _

Re: [Xen-devel] RTDS with extra time issue

2018-02-09 Thread Dario Faggioli
l actually be. Perhaps Meng has some more ideas on this as well. :-) Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description:

Re: [Xen-devel] RTDS with extra time issue

2018-02-09 Thread Dario Faggioli
If it does, then the scheduler is doing its job, and the issue may be somewhere else (e.g., something inside the guest may eat some of the budget, in such a way that not all of it is available when you actually need it). Let me know. Regards, Dario -- <> (Raistlin Majere) -

[Xen-devel] [PATCH 3/5] tools: libxl/xl: allow to get/set Credit1's vcpu_migration_delay

2018-02-22 Thread Dario Faggioli
Make it possible to get and set a (Credit1) scheduler's vCPU migration delay via the SCHEDOP sysctl, from both libxl and xl (no change needed in libxc). Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: Ian Jackson <ian.jack...@eu.citrix.com> Cc: Wei Liu <wei.l...@citrix.

[Xen-devel] [PATCH 0/5] xen/tools: sched: Credit1: improve handling of vCPU migration delay

2018-02-22 Thread Dario Faggioli
/fdario/xen/tree/rel/sched/credit/vcpu_migr_delay_percpool Travis is red, but that seems to me to be related to ARM infra issues, rather than to problem with the series itself: https://travis-ci.org/fdario/xen/builds/344835704 Regards, Dario --- Dario Faggioli (5): xen: sched/credit: convert

Re: [Xen-devel] RTDS with extra time issue

2018-02-22 Thread Dario Faggioli
On Tue, 2018-02-20 at 13:34 +0200, Andrii Anisov wrote: > Hello Dario, > Hi, > On 16.02.18 20:37, Dario Faggioli wrote: > > And in any case, is it, in its turn (I mean the > > workload running in DomR) a synthetic real-time load, or is it a > > real > > real

[Xen-devel] [PATCH 4/5] tools: xenpm: continue to support {set, get}-vcpu-migration-delay

2018-02-22 Thread Dario Faggioli
ard compatibility, and their use is deprecated, in favour of 'xl sched-credit -s [-c ] -m '. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: Ian Jackson <ian.jack...@eu.citrix.com> Cc: Wei Liu <wei.l...@citrix.com> Cc: George Dunlap <george.dun...@eu.citrix.com> Cc:

[Xen-devel] [PATCH 2/5] xen: sched/credit1: make vcpu_migration_delay per-cpupool

2018-02-22 Thread Dario Faggioli
operation, when the migration delay of a scheduler is zero (as it is, by default), in __csched_vcpu_is_cache_hot(). Finally, note that, from this commit on, using `xenpm {set,get}-vcpu-migration-delay' is not effective any longer. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: George

[Xen-devel] [PATCH 5/5] xen/libxc: suppress direct access to Credit1's migration delay

2018-02-22 Thread Dario Faggioli
Removes special purpose access to Credit1 vCPU migration delay parameter. This fixes a build breakage, occuring when Xen is configured with SCHED_CREDIT=n. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Ian Jackson <ian.jack...@eu.citrix.com> Wei Liu <wei.l...@citrix.com

[Xen-devel] [PATCH 1/5] xen: sched/credit: convert scheduling parameter to s_time_t when set

2018-02-22 Thread Dario Faggioli
to s_time_t), ncpus is moved up a bit, for better packing. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: George Dunlap <george.dun...@eu.citrix.com> Cc: Andrew Cooper <andrew.coop...@citrix.com> --- xen/common/sched_credit.c | 56 ++-

Re: [Xen-devel] [PATCH 5/5] xen/libxc: suppress direct access to Credit1's migration delay

2018-02-22 Thread Dario Faggioli
On Thu, 2018-02-22 at 17:30 +, Wei Liu wrote: > On Thu, Feb 22, 2018 at 05:55:32PM +0100, Dario Faggioli wrote: > > Removes special purpose access to Credit1 vCPU > > migration delay parameter. > > > > This fixes a build breakage, occuring when Xen > >

Re: [Xen-devel] [PATCH] tools: libxenstat: fix format string overflow

2018-02-16 Thread Dario Faggioli
On Fri, 2018-02-16 at 17:44 +, Wei Liu wrote: > On Fri, Feb 16, 2018 at 06:36:51PM +0100, Dario Faggioli wrote: > > > > --- a/tools/xenstat/libxenstat/src/xenstat_linux.c > > +++ b/tools/xenstat/libxenstat/src/xenstat_linux.c > > @@ -69,18 +69,20 @@ void getBri

Re: [Xen-devel] [PATCH] tools: libxenstat: fix format string overflow

2018-02-16 Thread Dario Faggioli
On Fri, 2018-02-16 at 17:44 +, Wei Liu wrote: > On Fri, Feb 16, 2018 at 06:36:51PM +0100, Dario Faggioli wrote: > > > > --- a/tools/xenstat/libxenstat/src/xenstat_linux.c > > +++ b/tools/xenstat/libxenstat/src/xenstat_linux.c > > @@ -69,18 +69,20 @@ void getBri

[Xen-devel] [PATCH v2] tools: libxenstat: fix format string overflow

2018-02-16 Thread Dario Faggioli
de->d_name); ^~ src/xenstat_linux.c:78:5: note: ‘sprintf’ output between 23 and 278 bytes into a destination of size 256 sprintf(tmp, "/sys/class/net/%s/bridge", de->d_name); ^~~~ Fix by making the buffer bigger. Sig

[Xen-devel] [PATCH] tools: libxenstat: fix format string overflow

2018-02-16 Thread Dario Faggioli
de->d_name); ^~ src/xenstat_linux.c:78:5: note: ‘sprintf’ output between 23 and 278 bytes into a destination of size 256 sprintf(tmp, "/sys/class/net/%s/bridge", de->d_name); ^~~~ Fix using asprintf(). Signed-off-

Re: [Xen-devel] [PATCH] tools: libxenstat: fix format string overflow

2018-02-16 Thread Dario Faggioli
On Fri, 2018-02-16 at 17:58 +, Wei Liu wrote: > On Fri, Feb 16, 2018 at 06:55:08PM +0100, Dario Faggioli wrote: > > On Fri, 2018-02-16 at 17:44 +, Wei Liu wrote: > > > > > Right! And what do I do if it fails, 'continue' the while(), I > > guess? > >

Re: [Xen-devel] RTDS with extra time issue

2018-02-16 Thread Dario Faggioli
ed to deal with a timer or whatever from DomY, then I agree it's not fair to charge DomX for that. But, OTOH, if we are in Xen because DomX itself called an hypercall, then it is indeed ok to charge DomX. And note that this does not have much to do with how schedule() gets called. :-) Regards, Dar

[Xen-devel] [PATCH v2 3/5] tools: libxl/xl: allow to get/set Credit1's vcpu_migration_delay

2018-02-23 Thread Dario Faggioli
Make it possible to get and set a (Credit1) scheduler's vCPU migration delay via the SCHEDOP sysctl, from both libxl and xl (no change needed in libxc). Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: Ian Jackson <ian.jack...@eu.citrix.com> Cc: Wei Liu <wei.l...@citrix.

[Xen-devel] [PATCH v2 5/5] xen/libxc: suppress direct access to Credit1's migration delay

2018-02-23 Thread Dario Faggioli
Removes special purpose access to Credit1 vCPU migration delay parameter. This fixes a build breakage, occuring when Xen is configured with SCHED_CREDIT=n. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> Acked-by: Wei Liu <wei.l...@citrix.com> --- Cc: Ian Jackson <ian.jack..

[Xen-devel] [PATCH v2 4/5] tools: xenpm: continue to support {set, get}-vcpu-migration-delay

2018-02-23 Thread Dario Faggioli
ard compatibility, and their use is deprecated, in favour of 'xl sched-credit -s [-c ] -m '. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> Acked-by: Wei Liu <wei.l...@citrix.com> --- Cc: Ian Jackson <ian.jack...@eu.citrix.com> Cc: Wei Liu <wei.l...@citrix.com> Cc: George Dunlap &

[Xen-devel] [PATCH v2 2/5] xen: sched/credit1: make vcpu_migration_delay per-cpupool

2018-02-23 Thread Dario Faggioli
operation, when the migration delay of a scheduler is zero (as it is, by default), in __csched_vcpu_is_cache_hot(). Finally, note that, from this commit on, using `xenpm {set,get}-vcpu-migration-delay' is not effective any longer. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: George

[Xen-devel] [PATCH v2 1/5] xen: sched/credit: convert scheduling parameter to s_time_t when set

2018-02-23 Thread Dario Faggioli
to s_time_t), ncpus is moved up a bit, for better packing. Signed-off-by: Dario Faggioli <dfaggi...@suse.com> --- Cc: George Dunlap <george.dun...@eu.citrix.com> Cc: Andrew Cooper <andrew.coop...@citrix.com> --- xen/common/sched_credit.c | 56 ++-

[Xen-devel] [PATCH v2 0/5] xen/tools: sched: Credit1: improve handling of vCPU migration delay

2018-02-23 Thread Dario Faggioli
/vcpu_migr_delay_percpool-v2 Regards, Dario --- Dario Faggioli (5): xen: sched/credit: convert scheduling parameter to s_time_t when set xen: sched/credit1: make vcpu_migration_delay per-cpupool tools: libxl/xl: allow to get/set Credit1's vcpu_migration_delay tools: xenpm: continue

Re: [Xen-devel] [PATCH 5/5] xen/libxc: suppress direct access to Credit1's migration delay

2018-02-23 Thread Dario Faggioli
set_vcpu_migration_delay; > > } u; > > }; > > Such removal imo needs to be accompanied by bumping of the > interface version. > Right! Sorry, I keep forgetting about that. Will do. Dario -- <> (Raistlin Majere) ---

Re: [Xen-devel] [PATCH 3/5] tools: libxl/xl: allow to get/set Credit1's vcpu_migration_delay

2018-02-23 Thread Dario Faggioli
On Thu, 2018-02-22 at 17:55 +0100, Dario Faggioli wrote: > --- a/tools/libxl/libxl_sched.c > +++ b/tools/libxl/libxl_sched.c > @@ -505,6 +507,9 @@ int main_sched_credit(int argc, char **argv) > ratelimit = strtol(optarg, NULL, 10); > opt_r = true; > bre

[Xen-devel] [PATCH] xen: cpupool: forbid to split cores among different pools

2018-08-20 Thread Dario Faggioli
, - outside of any pool. Signed-off-by: Dario Faggioli --- Cc: Juergen Gross --- xen/common/cpupool.c | 34 +- 1 file changed, 29 insertions(+), 5 deletions(-) diff --git a/xen/common/cpupool.c b/xen/common/cpupool.c index 1e8edcbd57..1e52fea5ac 100644 --- a/xen

[Xen-devel] [PATCH] tools: libxl/xl: run NUMA placement even when an hard-affinity is set

2018-08-17 Thread Dario Faggioli
aa67b97ed34279c43 ("xl.conf: Add global affinity masks"). In fact, without this commit, defining a global affinity mask would also mean disabling automatic placement, but that does not necessarily have to be the case (especially in large systems). Signed-off-by: Dario Faggioli --- Cc: Ian J

Re: [Xen-devel] [PATCH] xen: cpupool: forbid to split cores among different pools

2018-08-21 Thread Dario Faggioli
On Tue, 2018-08-21 at 10:25 +0200, Juergen Gross wrote: > On 20/08/18 18:43, Dario Faggioli wrote: > > > > Therefore, let's start allowing in a cpupool only cpus that have > > their > > SMT siblings, either: > > - in that same pool, > > - outside of a

Re: [Xen-devel] [PATCH] tools: libxl/xl: run NUMA placement even when an hard-affinity is set

2018-08-21 Thread Dario Faggioli
On Tue, 2018-08-21 at 11:25 +0100, Ian Jackson wrote: > Dario Faggioli writes ("Re: [Xen-devel] [PATCH] tools: libxl/xl: run > NUMA placement even when an hard-affinity is set"): > > On Mon, 2018-08-20 at 11:14 +0100, Wei Liu wrote: > > > You can use NDEBUG instead.

Re: [Xen-devel] L1TF, and future work

2018-08-24 Thread Dario Faggioli
, but expect me to reply back only in September. Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description: This is a digitally

[Xen-devel] [RFC PATCH v1 00/16] xen: sched: implement core-scheduling

2018-08-24 Thread Dario Faggioli
/xen.git rel/sched/core-scheduling-RFCv1 Any comment is more than welcome. Regards, Dario --- Dario Faggioli (16): xen: Credit1: count runnable vcpus, not running ones xen: Credit1: always steal from pcpus with runnable but not running vcpus xen: Credit1: do not always tickle

[Xen-devel] [RFC PATCH v1 01/16] xen: Credit1: count runnable vcpus, not running ones

2018-08-24 Thread Dario Faggioli
. This is going to be useful in following patches. Therefore, as long as just this commit is considered, there is no functional change (intended). Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 43 +-- 1 file changed, 13

[Xen-devel] [RFC PATCH v1 02/16] xen: Credit1: always steal from pcpus with runnable but not running vcpus

2018-08-24 Thread Dario Faggioli
sure that runnable vcpus are picked up from the runqueues of whatever pcpu they're in. We also make it more clear, while tracing, whether a cpu is being considered or skipped (and, if the latter, why) during load balancing. Signed-off-by: Dario Faggioli --- Cc: George Dunlap Cc: Ian Jackson Cc

[Xen-devel] [RFC PATCH v1 03/16] xen: Credit1: do not always tickle an idle pcpu

2018-08-24 Thread Dario Faggioli
where the vcpu wakes up, let's go through the tickling, even if it was idle. If there is no 'better' pcpu where to run it, we still prefer the one where it woke up, among others with similar characteristics. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 24

[Xen-devel] [RFC PATCH v1 07/16] xen: Credit1: reorg __runq_tickle() code a bit.

2018-08-24 Thread Dario Faggioli
-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 34 +- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/xen/common/sched_credit.c b/xen/common/sched_credit.c index af3b81d377..e218eb6986 100644 --- a/xen/common

[Xen-devel] [RFC PATCH v1 04/16] xen: sched: make the logic for tracking idle core generic.

2018-08-24 Thread Dario Faggioli
Introduced in 9bb9c73884d "xen: credit2: implement true SMT support", it was available only to Credit2. Move the functions to common headers so that also other schedulers can use them, to track the idleness of full cores. No functional change intended. Signed-off-by: Dario Faggio

[Xen-devel] [RFC PATCH v1 05/16] xen: Credit1: track fully idle cores.

2018-08-24 Thread Dario Faggioli
needs its own serialization. In fact, it can't be updated only with atomic cpumask operations, and hence its content would become totally unreliable, and not representative of the actual idleness status of the cores. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c

[Xen-devel] [RFC PATCH v1 10/16] xen: Credit1: support sched_smt_cosched in csched_schedule()

2018-08-24 Thread Dario Faggioli
other pcpu of core 1. It is, however, still possible that a vcpu from a random vcpu would be run on a non fully idle core, as a consequence of load balancing, which will be dealt with in a later commit. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 63

[Xen-devel] [RFC PATCH v1 08/16] xen: Credit1: reorg csched_schedule() code a bit.

2018-08-24 Thread Dario Faggioli
This is pure code refactoring, with the purpose of isolating code movement. No functional change intended. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 29 ++--- 1 file changed, 14 insertions(+), 15 deletions(-) diff --git a/xen

[Xen-devel] [PATCH v1 1/5] xen: sched: null: refactor core around vcpu_deassign()

2018-08-24 Thread Dario Faggioli
vcpu_deassign() has only one caller: _vcpu_remove(). Let's consolidate the two functions into one. No functional change intended. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_null.c | 76 ++- 1 file changed, 35

[Xen-devel] [PATCH v1 0/5] Series short description

2018-08-24 Thread Dario Faggioli
, with this patch, it works, that would be awesome. I can do that myself, of course, and I will, after vacations, if you don't get round to it by then. :-D Thanks and Regards, Dario --- Dario Faggioli (5): xen: sched: null: refactor core around vcpu_deassign() xen: sched: null: don't assign down

[Xen-devel] [PATCH v1 2/5] xen: sched: null: don't assign down vcpus to pcpus

2018-08-24 Thread Dario Faggioli
If a pCPU has been/is being offlined, we don't want it to be neither assigned to any pCPU, nor in the wait list. So, avoid doing that while inserting or migrating vcpus. Signed-off-by: Dario Faggioli --- Cc: George Dunlap Cc: Stefano Stabellini Cc: Roger Pau Monne --- xen/common

[Xen-devel] [RFC PATCH v1 09/16] xen: Credit1: SMT-aware domain co-scheduling parameter and data structs

2018-08-24 Thread Dario Faggioli
enforce the fact that only vcpus of the same domain runs on each core, whatever the tracking shows (e.g., via `xl debug-keys r') is not valid, and should be ignored. Signed-off-by: Dario Faggioli --- Cc: Andrew Cooper Cc: George Dunlap Cc: Jan Beulich Cc: Wei Liu --- TODO: - (in this patch

[Xen-devel] [RFC PATCH v1 06/16] xen: Credit1: check for fully idle cores when tickling

2018-08-24 Thread Dario Faggioli
. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 51 +++-- 1 file changed, 31 insertions(+), 20 deletions(-) diff --git a/xen/common/sched_credit.c b/xen/common/sched_credit.c index c7ee85d56a..af3b81d377 100644

[Xen-devel] [RFC PATCH v1 12/16] xen: Credit1: support sched_smt_cosched in csched_runq_steal().

2018-08-24 Thread Dario Faggioli
If sched_smt_cosched is enabled, when poking at other pcpus' runqueues (for doing load balancing), we only consider the vcpus of the domain that is running on the core already. Unless the core is fully idle, in which case, we can pick up any vcpu. Signed-off-by: Dario Faggioli --- Cc: George

[Xen-devel] [RFC PATCH v1 13/16] xen: Credit1: sched_smt_cosched support in __csched_vcpu_is_migrateable().

2018-08-24 Thread Dario Faggioli
If SMT domain co-scheduling is enabled, we only migrate a vcpu to either a fully idle core, or to pcpus of cores where other vcpus of the same domain are running already. Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 17 ++--- 1 file changed

[Xen-devel] [RFC PATCH v1 11/16] xen: Credit1: support sched_smt_cosched in _csched_cpu_pick()

2018-08-24 Thread Dario Faggioli
of domain A on pcpus of cores where vcpus of another domains are running!). Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 25 - 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/xen/common/sched_credit.c b/xen/common

[Xen-devel] [RFC PATCH v1 16/16] xen/tools: tracing of Credit1 SMT domain co-scheduling support

2018-08-24 Thread Dario Faggioli
Introduce some new event, related to SMT-aware domain co-scheduling, in Credit1 code, and their handling and parsing in xenalyze. Signed-off-by: Dario Faggioli --- Cc: George Dunlap Cc: Ian Jackson Cc: Wei Liu --- TODO: - deal with xentrace_format. --- tools/xentrace/xenalyze.c | 74

[Xen-devel] [RFC PATCH v1 14/16] xen: Credit1: sched_smt_cosched support in __runq_tickle() for pinned vcpus.

2018-08-24 Thread Dario Faggioli
When a vcpu which is exclusively pinned to a pcpu wakes up, we only have one option: tickling that exact pcpu. If sched_smt_cosched is enabled, however, it only makes sense to do that if pcpu is in a core where vcpus of the domain are running already. Signed-off-by: Dario Faggioli --- Cc

[Xen-devel] [RFC PATCH v1 15/16] xen: Credit1: sched_smt_cosched support in __runq_tickle().

2018-08-24 Thread Dario Faggioli
a bit more refining, in order to actually behave as described above (see the 'TODO's in the sources). Signed-off-by: Dario Faggioli --- Cc: George Dunlap --- xen/common/sched_credit.c | 60 ++--- 1 file changed, 51 insertions(+), 9 deletions(-) diff --git

[Xen-devel] [PATCH v1 5/5] xen: sched: null: refactor the ASSERTs around vcpu_deassing()

2018-08-24 Thread Dario Faggioli
It is all the time that we call vcpu_deassing() that the vcpu _must_ be assigned to a pCPU, and hence that such pCPU can't be free. Therefore, move the ASSERT-s which check for these properties in that function, where they belong better. --- Cc: George Dunlap --- xen/common/sched_null.c |5

[Xen-devel] [PATCH v1 3/5] xen: sched: null: deassign offline vcpus from pcpus

2018-08-24 Thread Dario Faggioli
If a pCPU has been/is being offlined, we don't want it to be neither assigned to any pCPU, nor in the wait list. Therefore, when we detect that a vcpu is going offline, remove it from both places. Signed-off-by: Dario Faggioli --- Cc: George Dunlap Cc: Stefano Stabellini Cc: Roger Pau Monne

[Xen-devel] [PATCH v1 4/5] xen: sched: null: reassign vcpus to pcpus when they come back online

2018-08-24 Thread Dario Faggioli
wever, all this new logic is triggered only by the case of a vcpu coming online, so, basically, the overhead during normal operations is just an additional 'if()'. Signed-off-by: Dario Faggioli --- Cc: George Dunlap Cc: Stefano Stabellini Cc: Roger Pau Monne --- xen/common/sched_nul

Re: [Xen-devel] [PATCH] tools: libxl/xl: run NUMA placement even when an hard-affinity is set

2018-08-20 Thread Dario Faggioli
On Mon, 2018-08-20 at 11:14 +0100, Wei Liu wrote: > On Fri, Aug 17, 2018 at 07:03:03PM +0200, Dario Faggioli wrote: > > @@ -27,6 +27,8 @@ > > > > #include "_paths.h" > > > > +//#define DEBUG 1 > > + > > Stray changes here? > > Y

Re: [Xen-devel] automation: Creating a patchwork instance to improve pre-commit build testing

2018-07-24 Thread Dario Faggioli
r > website, so that the bot won't reply to you if you don't want it to. > +1 :-) Regards, Dario -- <> (Raistlin Majere) - Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signatu

Re: [Xen-devel] [PATCH 2/6] xen/sched: Use %*pb[l] instead of cpumask_scn{, list}printf()

2018-09-12 Thread Dario Faggioli
iting > > parts of the same line. > > > > No functional change. > > > > Signed-off-by: Andrew Cooper > > Acked-by: George Dunlap > Acked-by: Dario Faggioli As it's a clear improvement wrt current state of things, IMO. I agree with Jan that having to re

Re: [Xen-devel] [PATCH 5/6] xen/bitmap: Drop all bitmap_scn{, list}printf() infrastructure

2018-09-12 Thread Dario Faggioli
off-by: Andrew Cooper > Reviewed-by: Dario Faggioli Regards, Dario -- <> (Raistlin Majere) ----- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Software Engineer @ SUSE https://www.suse.com/ signature.asc Description: This is a

Re: [Xen-devel] [PATCH 5/5] formats: print time values as decimals

2018-09-12 Thread Dario Faggioli
rii Anisov > > Acked-by: George Dunlap > Not strictly necessary, I now. Still: Reviewed-by: Dario Faggioli Andrii, if you decide to make the patch that turns timestamps into secs.nsecs, can we also convert these to such format (again, in case we have MHz)? Or, in general

  1   2   3   >