From: David Gibson <david@gibson.dropbear.id.au>
To: peter.maydell@linaro.org
Cc: qemu-devel@nongnu.org, qemu-ppc@nongnu.org, groug@kaod.org,
surajjs@au1.ibm.com, mark.cave-ayland@ilande.co.uk,
David Gibson <david@gibson.dropbear.id.au>
Subject: [Qemu-devel] [PULL 04/24] spapr: register dummy ICPs later
Date: Fri, 2 Mar 2018 17:03:30 +1100 [thread overview]
Message-ID: <20180302060350.24330-5-david@gibson.dropbear.id.au> (raw)
In-Reply-To: <20180302060350.24330-1-david@gibson.dropbear.id.au>
From: Greg Kurz <groug@kaod.org>
Some older machine types create more ICPs than needed. We hence
need to register up to xics_max_server_number() dummy ICPs to
accomodate the migration of these machine types.
Recent VSMT rework changed xics_max_server_number() to return
DIV_ROUND_UP(max_cpus * spapr->vsmt, smp_threads)
instead of
DIV_ROUND_UP(max_cpus * kvmppc_smt_threads(), smp_threads);
The change is okay but it requires spapr->vsmt to be set, which
isn't the case with the current code. This causes the formula to
return zero and we don't create dummy ICPs. This breaks migration
of older guests as reported here:
https://bugzilla.redhat.com/show_bug.cgi?id=1549087
The dummy ICP workaround doesn't really have a dependency on XICS
itself. But it does depend on proper VCPU id numbering and it must
be applied before creating vCPUs (ie, creating real ICPs). So this
patch moves the workaround to spapr_init_cpus(), which already
assumes VSMT to be set.
Fixes: 72194664c8a1 ("spapr: use spapr->vsmt to compute VCPU ids")
Reported-by: Lukas Doktor <ldoktor@redhat.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
---
hw/ppc/spapr.c | 24 ++++++++++++------------
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index d6fd0e666e..3f1c5c5133 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -183,7 +183,6 @@ static int xics_max_server_number(sPAPRMachineState *spapr)
static void xics_system_init(MachineState *machine, int nr_irqs, Error **errp)
{
sPAPRMachineState *spapr = SPAPR_MACHINE(machine);
- sPAPRMachineClass *smc = SPAPR_MACHINE_GET_CLASS(machine);
if (kvm_enabled()) {
if (machine_kernel_irqchip_allowed(machine) &&
@@ -205,17 +204,6 @@ static void xics_system_init(MachineState *machine, int nr_irqs, Error **errp)
return;
}
}
-
- if (smc->pre_2_10_has_unused_icps) {
- int i;
-
- for (i = 0; i < xics_max_server_number(spapr); i++) {
- /* Dummy entries get deregistered when real ICPState objects
- * are registered during CPU core hotplug.
- */
- pre_2_10_vmstate_register_dummy_icp(i);
- }
- }
}
static int spapr_fixup_cpu_smt_dt(void *fdt, int offset, PowerPCCPU *cpu,
@@ -2236,6 +2224,7 @@ static void spapr_init_cpus(sPAPRMachineState *spapr)
{
MachineState *machine = MACHINE(spapr);
MachineClass *mc = MACHINE_GET_CLASS(machine);
+ sPAPRMachineClass *smc = SPAPR_MACHINE_GET_CLASS(machine);
const char *type = spapr_get_cpu_core_type(machine->cpu_type);
const CPUArchIdList *possible_cpus;
int boot_cores_nr = smp_cpus / smp_threads;
@@ -2261,6 +2250,17 @@ static void spapr_init_cpus(sPAPRMachineState *spapr)
boot_cores_nr = possible_cpus->len;
}
+ if (smc->pre_2_10_has_unused_icps) {
+ int i;
+
+ for (i = 0; i < xics_max_server_number(spapr); i++) {
+ /* Dummy entries get deregistered when real ICPState objects
+ * are registered during CPU core hotplug.
+ */
+ pre_2_10_vmstate_register_dummy_icp(i);
+ }
+ }
+
for (i = 0; i < possible_cpus->len; i++) {
int core_id = i * smp_threads;
--
2.14.3
next prev parent reply other threads:[~2018-03-02 6:04 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-02 6:03 [Qemu-devel] [PULL 00/24] ppc-for-2.12 queue 20180302 David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 01/24] spapr: fix missing CPU core nodes in DT when running with TCG David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 02/24] ppc440: Add emulation of plb-pcix controller found in some 440 SoCs David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 03/24] ppc: Add aCube Sam460ex board David Gibson
2018-03-02 6:03 ` David Gibson [this message]
2018-03-02 6:03 ` [Qemu-devel] [PULL 05/24] spapr: harden code that depends on VSMT David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 06/24] macio: embed DBDMA device directly within macio David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 07/24] macio: move ESCC device within the macio device David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 08/24] heathrow: QOMify heathrow PIC David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 09/24] heathrow: convert to trace-events David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 10/24] heathrow: change heathrow_pic_init() to return the heathrow device David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 11/24] macio: move macio related structures and defines into separate macio.h file David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 12/24] mac_oldworld: use object link to pass heathrow PIC object to macio David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 13/24] openpic: move KVM-specific declarations into separate openpic_kvm.h file David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 14/24] openpic: move OpenPIC state and related definitions to openpic.h David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 15/24] mac_newworld: use object link to pass OpenPIC object to macio David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 16/24] macio: move setting of CUDA timebase frequency to macio_common_realize() David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 17/24] macio: remove macio_init() function David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 18/24] target/ppc: Check mask when setting cap_ppc_safe_indirect_branch David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 19/24] ppc/spapr-caps: Add support for custom spapr_capabilities David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 20/24] ppc/spapr-caps: Convert cap-cfpc to custom spapr-cap David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 21/24] ppc/spapr-caps: Convert cap-sbbc " David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 22/24] ppc/spapr-caps: Convert cap-ibs " David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 23/24] ppc/spapr-caps: Define the pseries-2.12-sxxm machine type David Gibson
2018-03-02 6:03 ` [Qemu-devel] [PULL 24/24] hw/ppc/spapr, e500: Use new property "stdout-path" for boot console David Gibson
2018-03-02 6:26 ` [Qemu-devel] [PULL 00/24] ppc-for-2.12 queue 20180302 no-reply
2018-03-02 14:26 ` Peter Maydell
2018-03-02 17:55 ` [Qemu-devel] [Qemu-ppc] " BALATON Zoltan
2018-03-02 21:51 ` BALATON Zoltan
2018-03-04 23:55 ` [Qemu-devel] " David Gibson
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=20180302060350.24330-5-david@gibson.dropbear.id.au \
--to=david@gibson.dropbear.id.au \
--cc=groug@kaod.org \
--cc=mark.cave-ayland@ilande.co.uk \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=surajjs@au1.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).