From: Peter Maydell <peter.maydell@linaro.org>
To: Nico Boehr <nrb@linux.ibm.com>
Cc: "Christian Borntraeger" <borntraeger@linux.ibm.com>,
"Nina Schoetterl-Glausch" <nsg@linux.ibm.com>,
qemu-devel@nongnu.org, "Thomas Huth" <thuth@redhat.com>,
qemu-s390x@nongnu.org,
"Richard Henderson" <richard.henderson@linaro.org>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>,
"David Hildenbrand" <david@redhat.com>,
"Ilya Leoshkevich" <iii@linux.ibm.com>,
"Cornelia Huck" <cohuck@redhat.com>,
"Halil Pasic" <pasic@linux.ibm.com>,
"Eric Farman" <farman@linux.ibm.com>,
"Michael Mueller" <mimu@linux.ibm.com>
Subject: Re: [PATCH for-9.2 00/10] s390: Convert virtio-ccw, cpu to three-phase reset, and followup cleanup
Date: Fri, 30 Aug 2024 13:04:54 +0100 [thread overview]
Message-ID: <CAFEAcA-FG0V0=MNf6sszGW-_Z8PdNv5DWZbLyiF3CCyTt8NUxQ@mail.gmail.com> (raw)
In-Reply-To: <172501900133.6066.3071687086903215470@t14-nrb.local>
On Fri, 30 Aug 2024 at 12:56, Nico Boehr <nrb@linux.ibm.com> wrote:
>
> Quoting Peter Maydell (2024-08-30 12:01:47)
> > I ran overnight with none of the patchset applied, and it never
> > failed (as expected). Running with just the first virtio-ccw
> > patch does fall over fairly quickly. So something's up with
> > that patch, which is curious because that's the one I thought
> > was a straightforward conversion without any complications :-)
> >
> > I'll investigate further today, I have the beginnings of a
> > theory about what might be happening...
>
> Thanks for taking the time, Peter! Let me know when you have insights.
I think I've found the problem, I'm just testing it to see if it
does properly fix the intermittent. The issue is that before we
can convert a class to three-phase reset we need to have
already converted all its parent classes. TYPE_VIRTIO_CCW_DEVICE
is a subclass of TYPE_CCW_DEVICE, but I forgot to convert
TYPE_CCW_DEVICE to three-phase reset. The result is that the
reset in the parent class was not happening at all (presumably
leaving the TYPE_CCW_DEVICE in an invalid/unexpected state on
reboot). The fix is to add an extra patch at the start of the
series. Once I've tested this I'll send out a v2 of the series,
maybe also adding the cleanup RTH suggested in one of the later
patches. (Ironically, if we do that cleanup then the bug would
also go away, because old-style DeviceClass:reset methods will
get called in exactly the same way as if they were registered
as phases.hold methods, rather than being specially handled...)
commit ff9bc4c8910f636f20e9b6e91d63dd003408ce10
Author: Peter Maydell <peter.maydell@linaro.org>
Date: Fri Aug 30 12:50:03 2024 +0100
hw/s390/ccw-device: Convert to three-phase reset
Convert the TYPE_CCW_DEVICE to three-phase reset. This is a
device class which is subclassed, so it needs to be three-phase
before we can convert the subclass.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
diff --git a/hw/s390x/ccw-device.c b/hw/s390x/ccw-device.c
index a7d682e5af9..14c24e38904 100644
--- a/hw/s390x/ccw-device.c
+++ b/hw/s390x/ccw-device.c
@@ -44,9 +44,9 @@ static Property ccw_device_properties[] = {
DEFINE_PROP_END_OF_LIST(),
};
-static void ccw_device_reset(DeviceState *d)
+static void ccw_device_reset_hold(Object *obj, ResetType type)
{
- CcwDevice *ccw_dev = CCW_DEVICE(d);
+ CcwDevice *ccw_dev = CCW_DEVICE(obj);
css_reset_sch(ccw_dev->sch);
}
@@ -55,11 +55,12 @@ static void ccw_device_class_init(ObjectClass
*klass, void *data)
{
DeviceClass *dc = DEVICE_CLASS(klass);
CCWDeviceClass *k = CCW_DEVICE_CLASS(klass);
+ ResettableClass *rc = RESETTABLE_CLASS(klass);
k->realize = ccw_device_realize;
k->refill_ids = ccw_device_refill_ids;
device_class_set_props(dc, ccw_device_properties);
- dc->reset = ccw_device_reset;
+ rc->phases.hold = ccw_device_reset_hold;
dc->bus_type = TYPE_VIRTUAL_CSS_BUS;
}
-- PMM
next prev parent reply other threads:[~2024-08-30 12:06 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-13 16:52 [PATCH for-9.2 00/10] s390: Convert virtio-ccw, cpu to three-phase reset, and followup cleanup Peter Maydell
2024-08-13 16:52 ` [PATCH for-9.2 01/10] hw/s390/virtio-ccw: Convert to three-phase reset Peter Maydell
2024-08-13 22:07 ` Richard Henderson
2024-08-14 10:43 ` Thomas Huth
2024-08-23 17:34 ` Nina Schoetterl-Glausch
2024-08-26 16:02 ` Philippe Mathieu-Daudé
2024-08-26 17:11 ` Peter Maydell
2024-08-13 16:52 ` [PATCH for-9.2 02/10] target/s390: Convert CPU to Resettable interface Peter Maydell
2024-08-13 23:03 ` Richard Henderson
2024-08-14 10:56 ` Thomas Huth
2024-08-23 17:44 ` Nina Schoetterl-Glausch
2024-08-23 20:32 ` Peter Maydell
2024-08-13 16:52 ` [PATCH for-9.2 03/10] hw: Remove device_class_set_parent_reset() Peter Maydell
2024-08-13 23:06 ` Richard Henderson
2024-08-26 16:03 ` Philippe Mathieu-Daudé
2024-08-13 16:52 ` [PATCH for-9.2 04/10] target/alpha, hppa: Remove unused parent_reset fields Peter Maydell
2024-08-13 23:08 ` Richard Henderson
2024-08-13 16:52 ` [PATCH for-9.2 05/10] hw/dma/xilinx_axidma: Use semicolon at end of statement, not comma Peter Maydell
2024-08-13 23:08 ` Richard Henderson
2024-08-14 10:55 ` Thomas Huth
2024-08-13 16:52 ` [PATCH for-9.2 06/10] hw/remote/message.c: Don't directly invoke DeviceClass:reset Peter Maydell
2024-08-13 23:10 ` Richard Henderson
2024-08-14 13:48 ` Philippe Mathieu-Daudé
2024-08-13 16:52 ` [PATCH for-9.2 07/10] hw: Define new device_class_set_legacy_reset() Peter Maydell
2024-08-13 23:12 ` Richard Henderson
2024-08-26 16:06 ` Philippe Mathieu-Daudé
2024-08-13 16:52 ` [PATCH for-9.2 08/10] hw: Use device_class_set_legacy_reset() instead of opencoding Peter Maydell
2024-08-13 23:14 ` Richard Henderson
2024-08-13 16:52 ` [PATCH for-9.2 09/10] hw: Rename DeviceClass::reset field to legacy_reset Peter Maydell
2024-08-13 23:15 ` Richard Henderson
2024-08-26 16:07 ` Philippe Mathieu-Daudé
2024-08-13 16:52 ` [PATCH for-9.2 10/10] hw: Remove device_phases_reset() Peter Maydell
2024-08-14 0:53 ` Richard Henderson
2024-08-14 13:20 ` Peter Maydell
2024-08-14 14:22 ` [PATCH for-9.2 00/10] s390: Convert virtio-ccw, cpu to three-phase reset, and followup cleanup Christian Borntraeger
2024-08-14 20:06 ` Peter Maydell
2024-08-22 10:34 ` Nina Schoetterl-Glausch
2024-08-26 12:08 ` Nico Boehr
2024-08-28 8:13 ` Nico Boehr
2024-08-28 15:46 ` Peter Maydell
2024-08-29 12:19 ` Nico Boehr
2024-08-29 13:09 ` Peter Maydell
2024-08-29 13:20 ` Cédric Le Goater
2024-08-29 13:26 ` Nico Boehr
2024-08-29 13:35 ` Peter Maydell
2024-08-29 14:44 ` Nico Boehr
2024-08-29 15:08 ` Peter Maydell
2024-08-29 15:53 ` Peter Maydell
2024-08-30 5:48 ` Nico Boehr
2024-08-30 10:01 ` Peter Maydell
2024-08-30 11:56 ` Nico Boehr
2024-08-30 12:04 ` Peter Maydell [this message]
2024-08-30 12:17 ` Peter Maydell
2024-08-30 12:21 ` Thomas Huth
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAFEAcA-FG0V0=MNf6sszGW-_Z8PdNv5DWZbLyiF3CCyTt8NUxQ@mail.gmail.com' \
--to=peter.maydell@linaro.org \
--cc=borntraeger@linux.ibm.com \
--cc=cohuck@redhat.com \
--cc=david@redhat.com \
--cc=farman@linux.ibm.com \
--cc=iii@linux.ibm.com \
--cc=mimu@linux.ibm.com \
--cc=nrb@linux.ibm.com \
--cc=nsg@linux.ibm.com \
--cc=pasic@linux.ibm.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-s390x@nongnu.org \
--cc=richard.henderson@linaro.org \
--cc=thuth@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).