!160 Automatically generate code patches with openeuler !47

From: @zhendongchen
Reviewed-by: @yorifang
Signed-off-by: @yorifang
This commit is contained in:
openeuler-ci-bot 2020-12-10 10:10:19 +08:00 committed by Gitee
commit ff04bde967
2 changed files with 37 additions and 1 deletions

View File

@ -0,0 +1,32 @@
From 38734e26ce3840d459da13607a9d46de24a15388 Mon Sep 17 00:00:00 2001
From: kevinZhu <zhukeqian94@163.com>
Date: Thu, 29 Oct 2020 19:24:48 +0800
Subject: [PATCH] Bugfix: hw/acpi: Use max_cpus instead of cpus when build PPTT
table
The field "cpus" is the initial number of CPU for guest, and the field "max_cpus"
is the max number of CPU after CPU hotplug. When building PPTT for guest, we
should take all CPUs into account, otherwise the "smp_sockets" is wrong.
Fixes: 7cfcd8c8a2fe ("build smt processor structure to support smt topology")
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
---
hw/acpi/aml-build.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/acpi/aml-build.c b/hw/acpi/aml-build.c
index 8a3b51c835..f01669df57 100644
--- a/hw/acpi/aml-build.c
+++ b/hw/acpi/aml-build.c
@@ -167,7 +167,7 @@ void build_pptt(GArray *table_data, BIOSLinker *linker, int possible_cpus)
struct offset_status offset;
const MachineState *ms = MACHINE(qdev_get_machine());
unsigned int smp_cores = ms->smp.cores;
- unsigned int smp_sockets = ms->smp.cpus / (smp_cores * ms->smp.threads);
+ unsigned int smp_sockets = ms->smp.max_cpus / (smp_cores * ms->smp.threads);
acpi_data_push(table_data, sizeof(AcpiTableHeader));
--
2.27.0

View File

@ -1,6 +1,6 @@
Name: qemu Name: qemu
Version: 4.1.0 Version: 4.1.0
Release: 33 Release: 34
Epoch: 2 Epoch: 2
Summary: QEMU is a generic and open source machine emulator and virtualizer Summary: QEMU is a generic and open source machine emulator and virtualizer
License: GPLv2 and BSD and MIT and CC-BY License: GPLv2 and BSD and MIT and CC-BY
@ -279,6 +279,7 @@ Patch0266: migration-multifd-fix-hangup-with-TLS-Multifd-due-to.patch
Patch0267: multifd-tls-fix-memoryleak-of-the-QIOChannelSocket-o.patch Patch0267: multifd-tls-fix-memoryleak-of-the-QIOChannelSocket-o.patch
Patch0268: net-remove-an-assert-call-in-eth_get_gso_type.patch Patch0268: net-remove-an-assert-call-in-eth_get_gso_type.patch
Patch0269: json-Fix-a-memleak-in-parse_pair.patch Patch0269: json-Fix-a-memleak-in-parse_pair.patch
Patch0270: Bugfix-hw-acpi-Use-max_cpus-instead-of-cpus-when-bui.patch
BuildRequires: flex BuildRequires: flex
BuildRequires: bison BuildRequires: bison
@ -625,6 +626,9 @@ getent passwd qemu >/dev/null || \
%endif %endif
%changelog %changelog
* Thu Oct 29 2020 Huawei Technologies Co., Ltd <alex.chen@huawei.com>
- Bugfix: hw/acpi: Use max_cpus instead of cpus when build PPTT table
* Fri Nov 13 2020 Huawei Technologies Co., Ltd <alex.chen@huawei.com> * Fri Nov 13 2020 Huawei Technologies Co., Ltd <alex.chen@huawei.com>
- json: Fix a memleak in parse_pair() - json: Fix a memleak in parse_pair()