[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-ppc] [Qemu-devel] [PULL 40/50] spapr_events: add support for p
From: |
Greg Kurz |
Subject: |
Re: [Qemu-ppc] [Qemu-devel] [PULL 40/50] spapr_events: add support for phb hotplug events |
Date: |
Fri, 1 Mar 2019 13:22:33 +0100 |
On Fri, 1 Mar 2019 11:49:30 +0100
Thomas Huth <address@hidden> wrote:
> On 01/03/2019 11.48, Greg Kurz wrote:
> > On Fri, 1 Mar 2019 11:30:18 +0100
> > David Hildenbrand <address@hidden> wrote:
> >
> >> On 01.03.19 02:31, Michael Roth wrote:
> >>> Quoting Thomas Huth (2019-02-28 12:40:52)
> >>>> On 26/02/2019 05.52, David Gibson wrote:
> >>>>> From: Michael Roth <address@hidden>
> >>>>>
> >>>>> Extend the existing EPOW event format we use for PCI
> >>>>> devices to emit PHB plug/unplug events.
> >>>>>
> >>>>> Signed-off-by: Michael Roth <address@hidden>
> >>>>> Reviewed-by: David Gibson <address@hidden>
> >>>>> Signed-off-by: Greg Kurz <address@hidden>
> >>>>> Message-Id: <address@hidden>
> >>>>> Signed-off-by: David Gibson <address@hidden>
> >>>>> ---
> >>>>> hw/ppc/spapr_events.c | 3 +++
> >>>>> 1 file changed, 3 insertions(+)
> >>>>>
> >>>>> diff --git a/hw/ppc/spapr_events.c b/hw/ppc/spapr_events.c
> >>>>> index b9c7ecb9e9..ab9a1f0063 100644
> >>>>> --- a/hw/ppc/spapr_events.c
> >>>>> +++ b/hw/ppc/spapr_events.c
> >>>>> @@ -526,6 +526,9 @@ static void spapr_hotplug_req_event(uint8_t hp_id,
> >>>>> uint8_t hp_action,
> >>>>> case SPAPR_DR_CONNECTOR_TYPE_CPU:
> >>>>> hp->hotplug_type = RTAS_LOG_V6_HP_TYPE_CPU;
> >>>>> break;
> >>>>> + case SPAPR_DR_CONNECTOR_TYPE_PHB:
> >>>>> + hp->hotplug_type = RTAS_LOG_V6_HP_TYPE_PHB;
> >>>>> + break;
> >>>>> default:
> >>>>> /* we shouldn't be signaling hotplug events for resources
> >>>>> * that don't support them
> >>>>
> >>>> I think this patch (or something else in this PULL request) broke CPU
> >>>> hot-plugging with older machine types:
> >>>>
> >>>> $ QTEST_QEMU_BINARY=ppc64-softmmu/qemu-system-ppc64
> >>>> ./tests/cpu-plug-test -m=slow
> >>>> /ppc64/cpu-plug/pseries-3.1/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-2.12-sxxm/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-3.0/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-2.10/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-2.11/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-2.12/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-2.9/device-add/2x3x1&maxcpus=6: OK
> >>>> /ppc64/cpu-plug/pseries-2.7/device-add/2x3x1&maxcpus=6: **
> >>>> ERROR:/home/thuth/devel/qemu/hw/ppc/spapr_events.c:313:rtas_event_log_to_source:
> >>>> assertion failed: (source->enabled)
> >>>> Broken pipe
> >>>> /home/thuth/devel/qemu/tests/libqtest.c:143: kill_qemu() detected QEMU
> >>>> death from signal 6 (Aborted) (core dumped)
> >>>> Aborted (core dumped)
> >>>>
> >>>> Could you please have a look?
> >>>
> >>> Bisected to:
> >>>
> >>> commit b8165118f52ce5ee88565d3cec83d30374efdc96
> >>> Author: David Hildenbrand <address@hidden>
> >>> Date: Mon Feb 18 10:21:58 2019 +0100
> >>>
> >>> spapr: support memory unplug for qtest
> >>>
> >>> Fake availability of OV5_HP_EVT, so we can test memory unplug in
> >>> qtest.
> >>>
> >>> Which makes sense since OV5_HP_EVT assumes that
> >>> spapr->spapr->use_hotplug_event_source == true, which isn't the default
> >>> for
> >>> 2.7 and below.
> >>>
> >>> If I revert that I think I hit the bug it was meant to fix:
> >>>
> >>> address@hidden:~/w/qemu-build3$ make V=1 check-qtest-ppc64
> >>> ...
> >>> PASS 1 device-plug-test /ppc64/device-plug/pci-unplug-request
> >>> PASS 2 device-plug-test /ppc64/device-plug/spapr-cpu-unplug-request
> >>> **
> >>>
> >>> ERROR:/home/mdroth/w/qemu3.git/tests/device-plug-test.c:28:device_del_finish:
> >>> assertion failed: (qdict_haskey(resp, "return"))
> >>> ERROR - Bail out!
> >>> ERROR:/home/mdroth/w/qemu3.git/tests/device-plug-test.c:28:device_del_finish:
> >>> assertion failed: (qdict_haskey(resp, "return"))
> >>> Aborted (core dumped)
> >>> /home/mdroth/w/qemu3.git/tests/Makefile.include:875: recipe for target
> >>> 'check-qtest-ppc64' failed
> >>> make: *** [check-qtest-ppc64] Error 1
> >>> address@hidden:~/w/qemu-build3$
> >>>
> >>> Which is probably due to this check in
> >>> spapr_machine_device_unplug_request():
> >>>
> >>> if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
> >>> if (spapr_ovec_test(sms->ov5_cas, OV5_HP_EVT)) {
> >>> spapr_memory_unplug_request(hotplug_dev, dev, errp);
> >>> } else {
> >>> /* NOTE: this means there is a window after guest reset,
> >>> prior to
> >>> * CAS negotiation, where unplug requests will fail due to the
> >>> * capability not being detected yet. This is a bit different
> >>> than
> >>> * the case with PCI unplug, where the events will be queued
> >>> and
> >>> * eventually handled by the guest after boot
> >>> */
> >>> error_setg(errp, "Memory hot unplug not supported for this
> >>> guest");
> >>> }
> >>>
> >>>
> >>>
> >>> This spapr-cpu-unplug-request test is failing because
> >>> spapr_ovec_test(sms->ov5_cas, OV5_HP_EVT) relies on the CAS-negotiated
> >>> OV5 bit,
> >>> which wouldn't have happened with qtest. If we want to make these tests
> >>> run in
> >>> this scenario we probably need a different approach than the original
> >>> patch.
> >>
> >> We could rip out the patch along with the spapr memory unplug test.
> >> However it feels like a step back to not have any memory unplug tests
> >> for QEMU at all.
> >>
> >> Any spapr experts here if we can work around this?
> >>
> >
> > Not sure about the expertise :) but I'm currently looking into it. As
> > you say, it would be unfortunate to drop a test because of that.
> >
>
> Could this work:
>
> diff --git a/hw/ppc/spapr_ovec.c b/hw/ppc/spapr_ovec.c
> --- a/hw/ppc/spapr_ovec.c
> +++ b/hw/ppc/spapr_ovec.c
> @@ -12,6 +12,7 @@
> */
>
> #include "qemu/osdep.h"
> +#include "hw/ppc/spapr.h"
> #include "hw/ppc/spapr_ovec.h"
> #include "qemu/bitmap.h"
> #include "exec/address-spaces.h"
> @@ -134,7 +135,9 @@ bool spapr_ovec_test(sPAPROptionVector *ov, long bitnr)
>
> /* support memory unplug for qtest */
> if (qtest_enabled() && bitnr == OV5_HP_EVT) {
> - return true;
> + sPAPRMachineState *spapr = SPAPR_MACHINE(qdev_get_machine());
> +
> + return spapr->use_hotplug_event_source;
> }
>
This patch definitely fixes make check but having to add yet another
qdev_get_machine() here is an indication that something is wrong.
An older machine type that doesn't have use_hotplug_event_source
shouldn't even bother about OV5_HP_EVT in the first place.
rtas_event_log_to_source() seems to assume that the machine has
an hotplug event source based on the fact that the guest advertised
it has support to use such a source. This might be true with a
regular guest because CAS would always clear the OV5_HP_EVT bit
if the machine didn't support it. Anyway, this is subtle and
thus fragile. If the machine doesn't know about hotplug event
source, we should step away from EVENT_CLASS_HOT_PLUG explicitly.
Something like:
---
diff --git a/hw/ppc/spapr_events.c b/hw/ppc/spapr_events.c
index ab9a1f0063d5..1a09dab6857d 100644
--- a/hw/ppc/spapr_events.c
+++ b/hw/ppc/spapr_events.c
@@ -307,11 +307,13 @@ rtas_event_log_to_source(sPAPRMachineState *spapr, int lo>
switch (log_type) {
case RTAS_LOG_TYPE_HOTPLUG:
- source = spapr_event_sources_get_source(spapr->event_sources,
- EVENT_CLASS_HOT_PLUG);
- if (spapr_ovec_test(spapr->ov5_cas, OV5_HP_EVT)) {
- g_assert(source->enabled);
- break;
+ if (spapr->use_hotplug_event_source) {
+ source = spapr_event_sources_get_source(spapr->event_sources,
+ EVENT_CLASS_HOT_PLUG);
+ if (spapr_ovec_test(spapr->ov5_cas, OV5_HP_EVT)) {
+ g_assert(source->enabled);
+ break;
+ }
}
/* fall back to epow for legacy hotplug interrupt source */
case RTAS_LOG_TYPE_EPOW:
---
> return test_bit(bitnr, ov->bitmap) ? true : false;
>
>
> ?
>
> Thomas