From: Cornelia Huck <cornelia.huck@de.ibm.com>
To: peter.maydell@linaro.org
Cc: borntraeger@de.ibm.com, agraf@suse.de, jfrei@linux.vnet.ibm.com,
qemu-devel@nongnu.org, Yi Min Zhao <zyimin@linux.vnet.ibm.com>,
Cornelia Huck <cornelia.huck@de.ibm.com>
Subject: [Qemu-devel] [PULL 25/25] s390x/pci: make hot-unplug handler smoother
Date: Mon, 11 Jul 2016 10:09:12 +0200 [thread overview]
Message-ID: <20160711080912.13947-26-cornelia.huck@de.ibm.com> (raw)
In-Reply-To: <20160711080912.13947-1-cornelia.huck@de.ibm.com>
From: Yi Min Zhao <zyimin@linux.vnet.ibm.com>
The current implementation of hot-unplug handler is abrupt. Any pci
operation will be just rejected if pci device is unconfigured. Thus a
pci device can not be reset or destroyed in a right, smooth and safe
way.
Improve this as follows:
- Notify the guest via a HP_EVENT_DECONFIGURE_REQUEST(0x303) event in
the unplug handler, giving it a chance to deconfigure the device via
sclp and allowing us to continue hot-unplug afterwards.
- Set up a timer that will generate the HP_EVENT_CONFIGURE_TO_STBRES
(0x304) event as before if the guest did not react after an adequate
time.
Signed-off-by: Yi Min Zhao <zyimin@linux.vnet.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
hw/s390x/s390-pci-bus.c | 35 ++++++++++++++++++++++++++++++++++-
hw/s390x/s390-pci-bus.h | 3 +++
2 files changed, 37 insertions(+), 1 deletion(-)
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index 630a826..640a4ea 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -192,6 +192,10 @@ void s390_pci_sclp_deconfigure(SCCB *sccb)
}
pbdev->state = ZPCI_FS_STANDBY;
rc = SCLP_RC_NORMAL_COMPLETION;
+
+ if (pbdev->release_timer) {
+ qdev_unplug(DEVICE(pbdev->pdev), NULL);
+ }
}
out:
psccb->header.response_code = cpu_to_be16(rc);
@@ -679,6 +683,23 @@ static void s390_pcihost_hot_plug(HotplugHandler *hotplug_dev,
}
}
+static void s390_pcihost_timer_cb(void *opaque)
+{
+ S390PCIBusDevice *pbdev = opaque;
+
+ if (pbdev->summary_ind) {
+ pci_dereg_irqs(pbdev);
+ }
+ if (pbdev->iommu_enabled) {
+ pci_dereg_ioat(pbdev);
+ }
+
+ pbdev->state = ZPCI_FS_STANDBY;
+ s390_pci_generate_plug_event(HP_EVENT_CONFIGURED_TO_STBRES,
+ pbdev->fh, pbdev->fid);
+ qdev_unplug(DEVICE(pbdev), NULL);
+}
+
static void s390_pcihost_hot_unplug(HotplugHandler *hotplug_dev,
DeviceState *dev, Error **errp)
{
@@ -712,8 +733,20 @@ static void s390_pcihost_hot_unplug(HotplugHandler *hotplug_dev,
case ZPCI_FS_STANDBY:
break;
default:
- s390_pci_generate_plug_event(HP_EVENT_CONFIGURED_TO_STBRES,
+ s390_pci_generate_plug_event(HP_EVENT_DECONFIGURE_REQUEST,
pbdev->fh, pbdev->fid);
+ pbdev->release_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL,
+ s390_pcihost_timer_cb,
+ pbdev);
+ timer_mod(pbdev->release_timer,
+ qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + HOT_UNPLUG_TIMEOUT);
+ return;
+ }
+
+ if (pbdev->release_timer && timer_pending(pbdev->release_timer)) {
+ timer_del(pbdev->release_timer);
+ timer_free(pbdev->release_timer);
+ pbdev->release_timer = NULL;
}
s390_pci_generate_plug_event(HP_EVENT_STANDBY_TO_RESERVED,
diff --git a/hw/s390x/s390-pci-bus.h b/hw/s390x/s390-pci-bus.h
index 71cdd7a..f1fbd3c 100644
--- a/hw/s390x/s390-pci-bus.h
+++ b/hw/s390x/s390-pci-bus.h
@@ -34,6 +34,7 @@
#define ZPCI_MAX_UID 0xffff
#define UID_UNDEFINED 0
#define UID_CHECKING_ENABLED 0x01
+#define HOT_UNPLUG_TIMEOUT (NANOSECONDS_PER_SECOND * 60 * 5)
#define S390_PCI_HOST_BRIDGE(obj) \
OBJECT_CHECK(S390pciState, (obj), TYPE_S390_PCI_HOST_BRIDGE)
@@ -44,6 +45,7 @@
#define HP_EVENT_TO_CONFIGURED 0x0301
#define HP_EVENT_RESERVED_TO_STANDBY 0x0302
+#define HP_EVENT_DECONFIGURE_REQUEST 0x0303
#define HP_EVENT_CONFIGURED_TO_STBRES 0x0304
#define HP_EVENT_STANDBY_TO_RESERVED 0x0308
@@ -283,6 +285,7 @@ typedef struct S390PCIBusDevice {
MemoryRegion iommu_mr;
IndAddr *summary_ind;
IndAddr *indicator;
+ QEMUTimer *release_timer;
} S390PCIBusDevice;
typedef struct S390PCIBus {
--
2.9.0
next prev parent reply other threads:[~2016-07-11 8:11 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-11 8:08 [Qemu-devel] [PULL 00/25] s390x patches for 2.7 Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 01/25] pc-bios/s390-ccw: Pass selected SCSI device to IPL Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 02/25] pc-bios/s390-ccw.img: rebuild image Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 03/25] s390x/ipl: Support IPL from selected SCSI device Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 04/25] s390x/ipl: fix reboots for migration from different bios Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 05/25] s390x/css: factor out some generic code from virtio_ccw_device_realize() Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 06/25] s390x/css: use define for "virtual-css-bridge" literal Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 07/25] s390x/css: Factor out virtual css bridge and bus Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 08/25] s390x/css: Unplug handler of virtual css bridge Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 09/25] s390x/pci: fix failures of dma map/unmap Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 10/25] s390x/pci: acceleration for getting S390pciState Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 11/25] s390x/pci: write fid in CLP_QUERY_PCI_FN Cornelia Huck
2016-07-11 8:08 ` [Qemu-devel] [PULL 12/25] s390x/pci: unify FH_ macros Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 13/25] s390x/pci: refactor s390_pci_find_dev_by_fh Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 14/25] s390x/pci: enforce zPCI state checking Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 15/25] s390x/pci: introduce S390PCIBus Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 16/25] s390x/pci: introduce S390PCIIOMMU Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 17/25] s390x/pci: introduce S390PCIBusDevice qdev Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 18/25] s390x/pci: enable uid-checking Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 19/25] s390x/pci: enable zpci hot-plug/hot-unplug Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 20/25] s390x/pci: add checkings in CLP_SET_PCI_FN Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 21/25] s390x/pci: refactor s390_pci_find_dev_by_idx Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 22/25] s390x/pci: refactor list_pci Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 23/25] s390x/pci: fix stpcifc_service_call Cornelia Huck
2016-07-11 8:09 ` [Qemu-devel] [PULL 24/25] s390x/pci: replace fid with idx in msg data of msix Cornelia Huck
2016-07-11 8:09 ` Cornelia Huck [this message]
2016-07-12 8:48 ` [Qemu-devel] [PULL 00/25] s390x patches for 2.7 Peter Maydell
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=20160711080912.13947-26-cornelia.huck@de.ibm.com \
--to=cornelia.huck@de.ibm.com \
--cc=agraf@suse.de \
--cc=borntraeger@de.ibm.com \
--cc=jfrei@linux.vnet.ibm.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=zyimin@linux.vnet.ibm.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).