rpms
/
qemu-kvm
7
0
Fork 0
Code Issues Pull Requests Releases Activity

Synchronization with CentOS 9 Stream Release 3 of 8.1.0 

Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
This commit is contained in:
Miroslav Rezanina 2023-10-16 12:42:05 +02:00
parent c76234dc30
commit 0d8d9bd342
35 changed files with 1848 additions and 80 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
0001-Initial-redhat-build.patch → 0004-Initial-redhat-build.patch
View File

@ -1,7 +1,7 @@
From 9dd5bb0af07c962833acf62f8d912c0472224482 Mon Sep 17 00:00:00 2001
From b8538e2099c040bfe7c98a3af1423abb30d91ab7 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 26 May 2021 10:56:02 +0200
Subject: [PATCH] Initial redhat build
Subject: Initial redhat build
This patch introduces redhat build structure in redhat subdirectory. In addition,
several issues are fixed in QEMU tree:
@ -187,12 +187,27 @@ Merged patches (8.1.0):
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
README.systemtap | 43 +++++++++++++++++++++++++
scripts/qemu-guest-agent/fsfreeze-hook | 2 +-
scripts/systemtap/conf.d/qemu_kvm.conf | 4 +++
scripts/systemtap/script.d/qemu_kvm.stp | 1 +
ui/vnc-auth-sasl.c | 2 +-
5 files changed, 50 insertions(+), 2 deletions(-)
.distro/Makefile | 100 +
.distro/Makefile.common | 42 +
.distro/README.tests | 39 +
.distro/modules-load.conf | 4 +
.distro/qemu-guest-agent.service | 1 -
.distro/qemu-kvm.spec.template | 4860 +++++++++++++++++++++++
.distro/rpminspect.yaml | 6 +-
.distro/scripts/extract_build_cmd.py | 12 +
.distro/scripts/process-patches.sh | 4 +
.gitignore | 1 +
README.systemtap | 43 +
scripts/qemu-guest-agent/fsfreeze-hook | 2 +-
scripts/systemtap/conf.d/qemu_kvm.conf | 4 +
scripts/systemtap/script.d/qemu_kvm.stp | 1 +
ui/vnc-auth-sasl.c | 2 +-
15 files changed, 5117 insertions(+), 4 deletions(-)
create mode 100644 .distro/Makefile
create mode 100644 .distro/Makefile.common
create mode 100644 .distro/README.tests
create mode 100644 .distro/modules-load.conf
create mode 100644 .distro/qemu-kvm.spec.template
create mode 100644 README.systemtap
create mode 100644 scripts/systemtap/conf.d/qemu_kvm.conf
create mode 100644 scripts/systemtap/script.d/qemu_kvm.stp
@ -289,3 +304,6 @@ index 47fdae5b21..2a950caa2a 100644
if (saslErr != SASL_OK) {
error_setg(errp, "Failed to initialize SASL auth: %s",
--
2.39.3

10
0002-Enable-disable-devices-for-RHEL.patch → 0005-Enable-disable-devices-for-RHEL.patch
View File

@ -1,7 +1,7 @@
From 2a8663a185aaafacac2414477c70d749be211d8e Mon Sep 17 00:00:00 2001
From ebbe3b0dc13b2f8ee7a1fcf9f6e687637a59ea35 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 7 Dec 2022 03:05:48 -0500
Subject: [PATCH] Enable/disable devices for RHEL
Subject: Enable/disable devices for RHEL
This commit adds all changes related to changes in supported devices.
@ -66,6 +66,7 @@ Merged patches (7.1.0):
Merged patches (8.1.0):
- 8173d2eaba Disable unwanted new devices
---
.distro/qemu-kvm.spec.template | 18 +--
.../aarch64-softmmu/aarch64-rh-devices.mak | 41 +++++++
.../ppc64-softmmu/ppc64-rh-devices.mak | 37 ++++++
configs/devices/rh-virtio.mak | 10 ++
@ -93,7 +94,7 @@ Merged patches (8.1.0):
target/s390x/cpu_models_sysemu.c | 3 +
target/s390x/kvm/kvm.c | 8 ++
tests/qtest/arm-cpu-features.c | 4 +
27 files changed, 311 insertions(+), 5 deletions(-)
28 files changed, 321 insertions(+), 13 deletions(-)
create mode 100644 configs/devices/aarch64-softmmu/aarch64-rh-devices.mak
create mode 100644 configs/devices/ppc64-softmmu/ppc64-rh-devices.mak
create mode 100644 configs/devices/rh-virtio.mak
@ -888,3 +889,6 @@ index 3fc33fc24d..e5e4007e9f 100644
sve_tests_default(qts, "max");
pauth_tests_default(qts, "max");
--
2.39.3

7
0003-Machine-type-related-general-changes.patch → 0006-Machine-type-related-general-changes.patch
View File

@ -1,7 +1,7 @@
From b4a38f074c6bcbb3e68177fe97f955a447864748 Mon Sep 17 00:00:00 2001
From 145fbe0a1526bb68a7bd7acaf5b41539faf5f04a Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 11 Jan 2019 09:54:45 +0100
Subject: [PATCH] Machine type related general changes
Subject: Machine type related general changes
This patch is first part of original "Add RHEL machine types" patch we
split to allow easier review. It contains changes not related to any
@ -680,3 +680,6 @@ index d54e8b1101..ae0f49b93c 100644
/* RAM / address space compat: */
bool gigabyte_align;
--
2.39.3

62
0004-Add-aarch64-machine-types.patch → 0007-Add-aarch64-machine-types.patch
View File

@ -1,11 +1,66 @@
From d5545d5287f04b69a052fa55c66d5ced5aa7b40e Mon Sep 17 00:00:00 2001
From f68fb44c79615701195203b2548135579b1fd596 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 12:53:31 +0200
Subject: [PATCH] Add aarch64 machine types
Subject: Add aarch64 machine types
Adding changes to add RHEL machine types for aarch64 architecture.
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
Rebase notes (6.1.0):
- Use CONFIG_TPM check when using TPM structures
- Add support for default_bus_bypass_iommu
- ea4c0b32d9 arm/virt: Register highmem and gic-version as class properties
- 895e1fa86a hw/arm/virt: Add 8.5 and 9.0 machine types and remove older ones
Rebase notes (7.0.0):
- Added dtb-kaslr-seed option
- Set no_tcg_lpa2 to true
Rebase notes (7.1.0):
- replace dtb_kaslr_seed by dtb_randomness
- Updated dtb_randomness comment
Rebase notes (7.2.0):
- Disabled cortex-a35
Rebase notes (8.0.0):
- Moved changed code from target/arm/helper.c to target/arm/arm-qmp-cmds.c
Rebase notes (8.1.0):
- Added setting default_nic
Merged patches (6.2.0):
- 9a3d4fde0e hw/arm/virt: Remove 9.0 machine type
- f7d04d6695 hw: arm: virt: Add hw_compat_rhel_8_5 to 8.5 machine type
Merged patches (7.0.0):
- 3b82be3dd3 redhat: virt-rhel8.5.0: Update machine type compatibility for QEMU 6.2.0 update
- c354a86c9b hw/arm/virt: Register "iommu" as a class property
- c1a2630dc9 hw/arm/virt: Register "its" as a class property
- 9d8c61dc93 hw/arm/virt: Rename default_bus_bypass_iommu
- a1d1b6eeb6 hw/arm/virt: Expose the 'RAS' option
- 47f8fe1b82 hw/arm/virt: Add 9.0 machine type and remove 8.5 one
- ed2346788f hw/arm/virt: Check no_tcg_its and minor style changes
Merged patches (7.0.0):
- f79b31bdef hw/arm/virt: Remove the dtb-kaslr-seed machine option
- b6fca85f4a hw/arm/virt: Fix missing initialization in instance/class_init()
Merged patches (7.1.0):
- ac97dd4f9f RHEL-only: AArch64: Drop unsupported CPU types
- e9c0a70664 target/arm: deprecate named CPU models
Merged patches (7.2.0):
- 0be2889fa2 Introduce upstream 7.0 compat changes (only applicable parts)
Merged patches (8.0.0):
- c1a21266d8 redhat: aarch64: add rhel9.2.0 virt machine type
- d97cd7c513 redhat: fix virt-rhel9.2.0 compat props
Merged patches (8.1.0):
- bd5d81d286 Add RHEL 9.2.0 compat structure (arm part)
- c07f666086 hw/arm/virt: Validate cluster and NUMA node boundary for RHEL machines
---
hw/arm/virt.c | 248 +++++++++++++++++++++++++++++++++++++++++-
include/hw/arm/virt.h | 8 ++
@ -367,3 +422,6 @@ index e1ddbea96b..81c2363a40 100644
void virt_acpi_setup(VirtMachineState *vms);
bool virt_is_acpi_enabled(VirtMachineState *vms);
--
2.39.3

7
0005-Add-ppc64-machine-types.patch → 0008-Add-ppc64-machine-types.patch
View File

@ -1,7 +1,7 @@
From 720842119b771090a08520e51f1772f234e1ee90 Mon Sep 17 00:00:00 2001
From bfe11e8d63f27ebc5bba4ca915e6e641af3a340b Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 13:27:13 +0200
Subject: [PATCH] Add ppc64 machine types
Subject: Add ppc64 machine types
Adding changes to add RHEL machine types for ppc64 architecture.
@ -539,3 +539,6 @@ index 6a4dd9c560..e0cd6fb1c7 100644
static inline int kvmppc_enable_hwrng(void)
{
return -1;
--
2.39.3

7
0006-Add-s390x-machine-types.patch → 0009-Add-s390x-machine-types.patch
View File

@ -1,7 +1,7 @@
From e30201e9f63350f9e7c44dd754942fe9e8dca785 Mon Sep 17 00:00:00 2001
From a3b5d9d0d0b59de4956ad88e5e855fe1a56ae215 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 13:47:32 +0200
Subject: [PATCH] Add s390x machine types
Subject: Add s390x machine types
Adding changes to add RHEL machine types for s390x architecture.
@ -285,3 +285,6 @@ index 87a4480c05..28c1b0486c 100644
/* check for unavailable features */
if (cpu_list_data->model) {
Object *obj;
--
2.39.3

33
0007-Add-x86_64-machine-types.patch → 0010-Add-x86_64-machine-types.patch
View File

@ -1,7 +1,7 @@
From 9bf78f1c31c8c06d957b2551dabd1150c0d42bfb Mon Sep 17 00:00:00 2001
From 70b66fab556a893f7f8b328f92bcd94c9b2978dd Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 13:10:31 +0200
Subject: [PATCH] Add x86_64 machine types
Subject: Add x86_64 machine types
Adding changes to add RHEL machine types for x86_64 architecture.
@ -66,7 +66,7 @@ Merged patches (8.1.0):
9 files changed, 557 insertions(+), 7 deletions(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 3109d5e0e0..8ff45cde20 100644
index 3109d5e0e0..4be9a7db82 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -410,6 +410,155 @@ GlobalProperty pc_compat_1_4[] = {
@ -88,7 +88,7 @@ index 3109d5e0e0..8ff45cde20 100644
+ { TYPE_X86_CPU, "host-phys-bits-limit", "48" },
+ { TYPE_X86_CPU, "vmx-entry-load-perf-global-ctrl", "off" },
+ { TYPE_X86_CPU, "vmx-exit-load-perf-global-ctrl", "off" },
+ /* bz 1508330 */
+ /* bz 1508330 */
+ { "vfio-pci", "x-no-geforce-quirks", "on" },
+ /* bz 1941397 */
+ { TYPE_X86_CPU, "kvm-asyncpf-int", "on" },
@ -191,27 +191,27 @@ index 3109d5e0e0..8ff45cde20 100644
+ * machine types irrespective of host.
+ */
+GlobalProperty pc_rhel_7_6_compat[] = {
+ /* pc_rhel_7_6_compat from pc_compat_3_0 */
+ /* pc_rhel_7_6_compat from pc_compat_3_0 */
+ { TYPE_X86_CPU, "x-hv-synic-kvm-only", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_3_0 */
+ /* pc_rhel_7_6_compat from pc_compat_3_0 */
+ { "Skylake-Server" "-" TYPE_X86_CPU, "pku", "off" },
+ /* pc_rhel_7_6_compat from pc_compat_3_0 */
+ /* pc_rhel_7_6_compat from pc_compat_3_0 */
+ { "Skylake-Server-IBRS" "-" TYPE_X86_CPU, "pku", "off" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { TYPE_X86_CPU, "x-migrate-smi-count", "off" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Skylake-Client" "-" TYPE_X86_CPU, "mpx", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Skylake-Client-IBRS" "-" TYPE_X86_CPU, "mpx", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Skylake-Server" "-" TYPE_X86_CPU, "mpx", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Skylake-Server-IBRS" "-" TYPE_X86_CPU, "mpx", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Cascadelake-Server" "-" TYPE_X86_CPU, "mpx", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Icelake-Client" "-" TYPE_X86_CPU, "mpx", "on" },
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ /* pc_rhel_7_6_compat from pc_compat_2_11 */
+ { "Icelake-Server" "-" TYPE_X86_CPU, "mpx", "on" },
+};
+const size_t pc_rhel_7_6_compat_len = G_N_ELEMENTS(pc_rhel_7_6_compat);
@ -907,3 +907,6 @@ index 78f1cf8186..ac954c9b06 100644
val = qtest_inb(qts, 0x505);
g_assert_cmpuint(val, ==, 3);
--
2.39.3

46
0008-Enable-make-check.patch → 0011-Enable-make-check.patch
View File

@ -1,13 +1,50 @@
From a869d740b1219361566895ea4429e2d78dfccfa8 Mon Sep 17 00:00:00 2001
From 301a52cd6a94e98e69b8beac2810e76f1cc47073 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 2 Sep 2020 09:39:41 +0200
Subject: [PATCH] Enable make check
Subject: Enable make check
Fixing tests after device disabling and machine types changes and enabling
make check run during build.
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
Rebase changes (6.1.0):
- removed unnecessary test changes
Rebase changes (6.2.0):
- new way of disabling bios-table-test
Rebase changes (7.0.0):
- Disable testing virtio-iommu-pci
- Rename default_bus_bypass_iommu property to default-bus-bypass-iommu
- Disable qtest-bios-table for aarch64
- Removed redhat chunks for boot-serial-test.c, cdrom-test.c and cpu-plug-test.c qtests
- Do not disable boot-order-test, prom-env-test and boot-serial-test qtests
- Use rhel machine type for new intel hda qtest
- Remove unnecessary changes in iotest 051
- Remove changes in bios-tables-test.c and prom-env-test.c qtests
Rebase changes (7.1.0):
- Disable bcm2835-dma-test (added upstream)
Rebase changes (8.0.0):
- Removed chunks for disabling bios-table-test (protected upstream)
- Disable new qemu-iotests execution
- Revert change in tco qtest (blocking test run)
Rebase changes (8.1.0):
- Do not disable device-plug-test for s390x
Merged patches (6.1.0):
- 2f129df7d3 redhat: Enable the 'test-block-iothread' test again
Merged patches (7.1.0):
- 64d736640e RHEL-only: tests/avocado: Switch aarch64 tests from a53 to a57
Merged patches (8.1.0):
- f468163234 iotests: Use alternative CPU type that is not deprecated in RHEL
---
.distro/qemu-kvm.spec.template | 4 ++--
tests/avocado/replay_kernel.py | 2 +-
tests/avocado/reverse_debugging.py | 2 +-
tests/avocado/tcg_plugins.py | 6 ++---
@ -22,7 +59,7 @@ Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
tests/qtest/tco-test.c | 2 +-
tests/qtest/usb-hcd-xhci-test.c | 4 ++++
tests/qtest/virtio-net-failover.c | 1 +
14 files changed, 36 insertions(+), 29 deletions(-)
15 files changed, 38 insertions(+), 31 deletions(-)
diff --git a/tests/avocado/replay_kernel.py b/tests/avocado/replay_kernel.py
index 79c607b0e7..c8827624f8 100644
@ -270,3 +307,6 @@ index 4a809590bf..1bf3fa641c 100644
"-device pcie-root-port,id=root0,addr=0x1,bus=pcie.0,chassis=1 " \
"-device pcie-root-port,id=root1,addr=0x2,bus=pcie.0,chassis=2 "
--
2.39.3

7
0009-vfio-cap-number-of-devices-that-can-be-assigned.patch → 0012-vfio-cap-number-of-devices-that-can-be-assigned.patch
View File

@ -1,7 +1,7 @@
From 2198cdd64f9317094a47fa7418e8aac28ab5466e Mon Sep 17 00:00:00 2001
From 6e5c0cde105d9250a6c46c06e44c4066cb804bf4 Mon Sep 17 00:00:00 2001
From: Bandan Das <bsd@redhat.com>
Date: Tue, 3 Dec 2013 20:05:13 +0100
Subject: [PATCH] vfio: cap number of devices that can be assigned
Subject: vfio: cap number of devices that can be assigned
RH-Author: Bandan Das <bsd@redhat.com>
Message-id: <1386101113-31560-3-git-send-email-bsd@redhat.com>
@ -101,3 +101,6 @@ index a2771b9ff3..a69af7bdde 100644
uint32_t vendor_id;
uint32_t device_id;
uint32_t sub_vendor_id;
--
2.39.3

7
0010-Add-support-statement-to-help-output.patch → 0013-Add-support-statement-to-help-output.patch
View File

@ -1,7 +1,7 @@
From 2be126e8077c079a3fad745384e205dcaf091082 Mon Sep 17 00:00:00 2001
From b6d7732d3fb071c96a86d8e98d8ef9f73932deaf Mon Sep 17 00:00:00 2001
From: Eduardo Habkost <ehabkost@redhat.com>
Date: Wed, 4 Dec 2013 18:53:17 +0100
Subject: [PATCH] Add support statement to -help output
Subject: Add support statement to -help output
RH-Author: Eduardo Habkost <ehabkost@redhat.com>
Message-id: <1386183197-27761-1-git-send-email-ehabkost@redhat.com>
@ -50,3 +50,6 @@ index b0b96f67fa..e24c74771b 100644
exit(exitcode);
}
--
2.39.3

7
0011-Use-qemu-kvm-in-documentation-instead-of-qemu-system.patch → 0014-Use-qemu-kvm-in-documentation-instead-of-qemu-system.patch
View File

@ -1,7 +1,7 @@
From f8bd6fa0cc6f2d8517b9333f5b4f0b5ee47c30c0 Mon Sep 17 00:00:00 2001
From 9f13719399c2b07b3fe8d691134363075ccd8a32 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 8 Jul 2020 08:35:50 +0200
Subject: [PATCH] Use qemu-kvm in documentation instead of qemu-system-<arch>
Subject: Use qemu-kvm in documentation instead of qemu-system-<arch>
Patchwork-id: 62380
O-Subject: [RHEV-7.1 qemu-kvm-rhev PATCHv4] Use qemu-kvm in documentation instead of qemu-system-i386
@ -56,3 +56,6 @@ index 29b98c3d4c..75a6396ed6 100644
``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]``
Establish a vhost-vdpa netdev.
--
2.39.3

12
0012-qcow2-Deprecation-warning-when-opening-v2-images-rw.patch → 0015-qcow2-Deprecation-warning-when-opening-v2-images-rw.patch
View File

@ -1,7 +1,7 @@
From 9e16cb3d2facc8fb20d502d1c94547141555cb32 Mon Sep 17 00:00:00 2001
From 28f0f014754e68ca5e5e00ba0dfbb38e3f60e77b Mon Sep 17 00:00:00 2001
From: Kevin Wolf <kwolf@redhat.com>
Date: Fri, 20 Aug 2021 18:25:12 +0200
Subject: [PATCH] qcow2: Deprecation warning when opening v2 images rw
Subject: qcow2: Deprecation warning when opening v2 images rw
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
@ -33,6 +33,11 @@ uses it this way for another warning.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
patch_name: kvm-qcow2-Deprecation-warning-when-opening-v2-images-rw.patch
present_in_specfile: true
location_in_specfile: 116
---
Rebase notes (6.1.0):
- Replace bs->read_only with bdrv_is_read_only
---
block/qcow2.c | 6 ++++++
tests/qemu-iotests/common.filter | 1 +
@ -67,3 +72,6 @@ index fc3c64bcb8..4b238954d5 100644
-e $'s#\r##' # QEMU monitor uses \r\n line endings
}
--
2.39.3

9
kvm-vdpa-use-first-queue-SVQ-state-for-CVQ-default.patch → 0016-vdpa-use-first-queue-SVQ-state-for-CVQ-default.patch
View File

@ -1,7 +1,7 @@
From 6048ff87666e1c1f023b6c39b25e26e878f79869 Mon Sep 17 00:00:00 2001
From 1b07788415a3d9729a3b11fc14d1d56ce5e6c888 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Thu, 10 Aug 2023 16:08:18 +0200
Subject: [PATCH] vdpa: use first queue SVQ state for CVQ default
Subject: vdpa: use first queue SVQ state for CVQ default
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
@ -26,6 +26,8 @@ Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Patch-name: kvm-vdpa-use-first-queue-SVQ-state-for-CVQ-default.patch
Patch-id: 189
Patch-present-in-specfile: True
---
net/vhost-vdpa.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
@ -43,3 +45,6 @@ index 9795306742..a772540250 100644
s->vhost_vdpa.address_space_id = VHOST_VDPA_GUEST_PA_ASID;
if (s->vhost_vdpa.shadow_data) {
--
2.39.3

9
kvm-vdpa-export-vhost_vdpa_set_vring_ready.patch → 0017-vdpa-export-vhost_vdpa_set_vring_ready.patch
View File

@ -1,7 +1,7 @@
From 83f01998db0d32a738a0f0beb4480d762fa6e025 Mon Sep 17 00:00:00 2001
From 90235be168d252feb40376ea5a3cc67fcfe097af Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Wed, 25 Jan 2023 08:47:34 +0100
Subject: [PATCH] vdpa: export vhost_vdpa_set_vring_ready
Subject: vdpa: export vhost_vdpa_set_vring_ready
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
@ -27,6 +27,8 @@ Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Patch-name: kvm-vdpa-export-vhost_vdpa_set_vring_ready.patch
Patch-id: 190
Patch-present-in-specfile: True
---
hw/virtio/trace-events | 2 +-
hw/virtio/vhost-vdpa.c | 25 +++++++++++++------------
@ -102,3 +104,6 @@ index e64bfc7f98..5407d54fd7 100644
int vhost_vdpa_dma_map(struct vhost_vdpa *v, uint32_t asid, hwaddr iova,
hwaddr size, void *vaddr, bool readonly);
--
2.39.3

9
kvm-vdpa-rename-vhost_vdpa_net_load-to-vhost_vdpa_net_cv.patch → 0018-vdpa-rename-vhost_vdpa_net_load-to-vhost_vdpa_net_cv.patch
View File

@ -1,7 +1,7 @@
From 91076a2975ed193b277c9b96f76ed1f86242c9ed Mon Sep 17 00:00:00 2001
From 772d982f7aa1101dcd873b2f7d800d9f841d3171 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Wed, 9 Aug 2023 18:07:26 +0200
Subject: [PATCH] vdpa: rename vhost_vdpa_net_load to vhost_vdpa_net_cvq_load
Subject: vdpa: rename vhost_vdpa_net_load to vhost_vdpa_net_cvq_load
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
@ -19,6 +19,8 @@ Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Patch-name: kvm-vdpa-rename-vhost_vdpa_net_load-to-vhost_vdpa_net_cv.patch
Patch-id: 191
Patch-present-in-specfile: True
---
net/vhost-vdpa.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
@ -45,3 +47,6 @@ index a772540250..9251351b4b 100644
.stop = vhost_vdpa_net_cvq_stop,
.cleanup = vhost_vdpa_cleanup,
.has_vnet_hdr = vhost_vdpa_has_vnet_hdr,
--
2.39.3

10
0016-vdpa-move-vhost_vdpa_set_vring_ready-to-the-caller.patch → 0019-vdpa-move-vhost_vdpa_set_vring_ready-to-the-caller.patch
View File

@ -1,7 +1,7 @@
From 11412870b7bfb3ebb63136f4cc409afb0760f08b Mon Sep 17 00:00:00 2001
From aa519f65c777fd7dbb1236a4f4db73f90e51660c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Thu, 10 Aug 2023 11:27:28 +0200
Subject: [PATCH] vdpa: move vhost_vdpa_set_vring_ready to the caller
Subject: vdpa: move vhost_vdpa_set_vring_ready to the caller
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
@ -23,6 +23,9 @@ the function is already static so it would only add an empty
vhost_vdpa_net_data_load stub.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
---
Rebase notes (8.1.0):
- Moved additional upstream calls to conditional block
---
hw/virtio/vdpa-dev.c | 3 +++
hw/virtio/vhost-vdpa.c | 3 ---
@ -141,3 +144,6 @@ index 9251351b4b..3bf60f9431 100644
}
return 0;
--
2.39.3

9
kvm-vdpa-remove-net-cvq-migration-blocker.patch → 0020-vdpa-remove-net-cvq-migration-blocker.patch
View File

@ -1,7 +1,7 @@
From d6da22076b3f59c2a0f664da93fc838e4270f9fe Mon Sep 17 00:00:00 2001
From aab6b27d559afb03cac1e9f873a65abc540b04c8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
Date: Fri, 24 Mar 2023 13:28:15 +0100
Subject: [PATCH] vdpa: remove net cvq migration blocker
Subject: vdpa: remove net cvq migration blocker
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
@ -21,6 +21,8 @@ Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Patch-name: kvm-vdpa-remove-net-cvq-migration-blocker.patch
Patch-id: 193
Patch-present-in-specfile: True
---
net/vhost-vdpa.c | 12 ------------
1 file changed, 12 deletions(-)
@ -48,3 +50,6 @@ index 3bf60f9431..6bb56f7d94 100644
}
ret = vhost_vdpa_add(nc, (void *)&s->vhost_vdpa, queue_pair_index, nvqs);
if (ret) {
--
2.39.3

9
0018-Add-machine-types-compat-bits.patch → 0021-Add-machine-types-compat-bits.patch
View File

@ -1,7 +1,7 @@
From 00d4a5989a5f8437ccbb4a6d1ef186cb6cbe05aa Mon Sep 17 00:00:00 2001
From 4ee284aca9a6fb151f85f266b375a8c1d9213a28 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 28 Jun 2023 01:40:03 -0400
Subject: [PATCH] Add machine types compat bits.
Subject: Add machine types compat bits.
Adding upstream compat changes to rhel machine types.
@ -49,7 +49,7 @@ index 2e54c7ae7c..8e23b9263d 100644
/* hw_compat_rhel_9_2 from hw_compat_7_2 */
{ "e1000e", "migrate-timadj", "off" },
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 8ff45cde20..ea7cf2fa15 100644
index 4be9a7db82..173815cc7e 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -432,6 +432,12 @@ GlobalProperty pc_rhel_compat[] = {
@ -135,3 +135,6 @@ index 7b569fa254..eb6ff6b099 100644
extern GlobalProperty pc_rhel_9_2_compat[];
extern const size_t pc_rhel_9_2_compat_len;
--
2.39.3

7
0019-Fix-x86-machine-type-compatibility-for-qemu-kvm-8.1..patch → 0022-Fix-x86-machine-type-compatibility-for-qemu-kvm-8.1..patch
View File

@ -1,7 +1,7 @@
From 6a2356f6102cd8757744349911ada90c015a56b1 Mon Sep 17 00:00:00 2001
From 719e2ac147c64d3caa1b86fb94a4182e9ff43f1b Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Tue, 22 Aug 2023 14:36:29 +0200
Subject: [PATCH] Fix x86 machine type compatibility for qemu-kvm 8.1.0
Subject: Fix x86 machine type compatibility for qemu-kvm 8.1.0
Jira: https://issues.redhat.com/browse/RHEL-874
Upstream Status: RHEL-only
@ -56,3 +56,6 @@ index 0b8adb0adb..a205226381 100644
pc_rhel_9_3_compat_len);
compat_props_add(m->compat_props, hw_compat_rhel_9_2,
hw_compat_rhel_9_2_len);
--
2.39.3

73
kvm-file-posix-Check-bs-bl.zoned-for-zone-info.patch Normal file
View File

@ -0,0 +1,73 @@
From 42dac3d063e0f44e92211a4fbdaf1f0d64843562 Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Thu, 24 Aug 2023 17:53:41 +0200
Subject: [PATCH 10/13] file-posix: Check bs->bl.zoned for zone info
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 202: file-posix: Fix zone update in I/O error path
RH-Jira: RHEL-7360
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [2/5] d853b88d6b71b6f4fe1bb70ae2e3b6108cbdb8dc (hreitz/qemu-kvm-c-9-s)
Instead of checking bs->wps or bs->bl.zone_size for whether zone
information is present, check bs->bl.zoned. That is the flag that
raw_refresh_zoned_limits() reliably sets to indicate zone support. If
it is set to something other than BLK_Z_NONE, other values and objects
like bs->wps and bs->bl.zone_size must be non-null/zero and valid; if it
is not, we cannot rely on their validity.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230824155345.109765-3-hreitz@redhat.com>
Reviewed-by: Sam Li <faithilikerun@gmail.com>
(cherry picked from commit 4b5d80f3d02096a9bb1f651f6b3401ba40877159)
Downstream: Fixed typo in the last hunk ("bs->blk" changed to
"bs->blk"). This code is removed in the commit after the
next, still, should be fixed now that I noticed.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/file-posix.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 2b88b9eefa..9f14850e24 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -2455,9 +2455,10 @@ static int coroutine_fn raw_co_prw(BlockDriverState *bs, uint64_t offset,
if (fd_open(bs) < 0)
return -EIO;
#if defined(CONFIG_BLKZONED)
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) && bs->wps) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
qemu_co_mutex_lock(&bs->wps->colock);
- if (type & QEMU_AIO_ZONE_APPEND && bs->bl.zone_size) {
+ if (type & QEMU_AIO_ZONE_APPEND) {
int index = offset / bs->bl.zone_size;
offset = bs->wps->wp[index];
}
@@ -2508,8 +2509,8 @@ out:
{
BlockZoneWps *wps = bs->wps;
if (ret == 0) {
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND))
- && wps && bs->bl.zone_size) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
uint64_t *wp = &wps->wp[offset / bs->bl.zone_size];
if (!BDRV_ZT_IS_CONV(*wp)) {
if (type & QEMU_AIO_ZONE_APPEND) {
@@ -2529,7 +2530,8 @@ out:
}
}
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) && wps) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
qemu_co_mutex_unlock(&wps->colock);
}
}
--
2.39.3

96
kvm-file-posix-Clear-bs-bl.zoned-on-error.patch Normal file
View File

@ -0,0 +1,96 @@
From 1eba7b686a41c71f24f60f8adcd51ddf1882686d Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Thu, 24 Aug 2023 17:53:40 +0200
Subject: [PATCH 09/13] file-posix: Clear bs->bl.zoned on error
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 202: file-posix: Fix zone update in I/O error path
RH-Jira: RHEL-7360
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [1/5] 3e2419b64b35afdfa45f70c5eb61478321de5233 (hreitz/qemu-kvm-c-9-s)
bs->bl.zoned is what indicates whether the zone information is present
and valid; it is the only thing that raw_refresh_zoned_limits() sets if
CONFIG_BLKZONED is not defined, and it is also the only thing that it
sets if CONFIG_BLKZONED is defined, but there are no zones.
Make sure that it is always set to BLK_Z_NONE if there is an error
anywhere in raw_refresh_zoned_limits() so that we do not accidentally
announce zones while our information is incomplete or invalid.
This also fixes a memory leak in the last error path in
raw_refresh_zoned_limits().
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230824155345.109765-2-hreitz@redhat.com>
Reviewed-by: Sam Li <faithilikerun@gmail.com>
(cherry picked from commit 56d1a022a77ea2125564913665eeadf3e303a671)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/file-posix.c | 21 ++++++++++++---------
1 file changed, 12 insertions(+), 9 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index b16e9c21a1..2b88b9eefa 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -1412,11 +1412,9 @@ static void raw_refresh_zoned_limits(BlockDriverState *bs, struct stat *st,
BlockZoneModel zoned;
int ret;
- bs->bl.zoned = BLK_Z_NONE;
-
ret = get_sysfs_zoned_model(st, &zoned);
if (ret < 0 || zoned == BLK_Z_NONE) {
- return;
+ goto no_zoned;
}
bs->bl.zoned = zoned;
@@ -1437,10 +1435,10 @@ static void raw_refresh_zoned_limits(BlockDriverState *bs, struct stat *st,
if (ret < 0) {
error_setg_errno(errp, -ret, "Unable to read chunk_sectors "
"sysfs attribute");
- return;
+ goto no_zoned;
} else if (!ret) {
error_setg(errp, "Read 0 from chunk_sectors sysfs attribute");
- return;
+ goto no_zoned;
}
bs->bl.zone_size = ret << BDRV_SECTOR_BITS;
@@ -1448,10 +1446,10 @@ static void raw_refresh_zoned_limits(BlockDriverState *bs, struct stat *st,
if (ret < 0) {
error_setg_errno(errp, -ret, "Unable to read nr_zones "
"sysfs attribute");
- return;
+ goto no_zoned;
} else if (!ret) {
error_setg(errp, "Read 0 from nr_zones sysfs attribute");
- return;
+ goto no_zoned;
}
bs->bl.nr_zones = ret;
@@ -1472,10 +1470,15 @@ static void raw_refresh_zoned_limits(BlockDriverState *bs, struct stat *st,
ret = get_zones_wp(bs, s->fd, 0, bs->bl.nr_zones, 0);
if (ret < 0) {
error_setg_errno(errp, -ret, "report wps failed");
- bs->wps = NULL;
- return;
+ goto no_zoned;
}
qemu_co_mutex_init(&bs->wps->colock);
+ return;
+
+no_zoned:
+ bs->bl.zoned = BLK_Z_NONE;
+ g_free(bs->wps);
+ bs->wps = NULL;
}
#else /* !defined(CONFIG_BLKZONED) */
static void raw_refresh_zoned_limits(BlockDriverState *bs, struct stat *st,
--
2.39.3

42
kvm-file-posix-Fix-zone-update-in-I-O-error-path.patch Normal file
View File

@ -0,0 +1,42 @@
From fc9ca44d7df182a0a483d8376054850ba46e633d Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Thu, 24 Aug 2023 17:53:42 +0200
Subject: [PATCH 11/13] file-posix: Fix zone update in I/O error path
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 202: file-posix: Fix zone update in I/O error path
RH-Jira: RHEL-7360
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [3/5] 28a8540b2ac0c437f8afb90fd5566fc83bfcfdc4 (hreitz/qemu-kvm-c-9-s)
We must check that zone information is present before running
update_zones_wp().
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=2234374
Fixes: Coverity CID 1512459
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230824155345.109765-4-hreitz@redhat.com>
Reviewed-by: Sam Li <faithilikerun@gmail.com>
(cherry picked from commit deab5c9a4ed74f76a713008a42527762b30a7e84)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/file-posix.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 9f14850e24..b9d5e4741b 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -2525,7 +2525,8 @@ out:
}
}
} else {
- if (type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
update_zones_wp(bs, s->fd, 0, 1);
}
}
--
2.39.3

71
kvm-file-posix-Simplify-raw_co_prw-s-out-zone-code.patch Normal file
View File

@ -0,0 +1,71 @@
From fbf511a9f7272815fc6d7414735049689186c750 Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Thu, 24 Aug 2023 17:53:43 +0200
Subject: [PATCH 12/13] file-posix: Simplify raw_co_prw's 'out' zone code
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 202: file-posix: Fix zone update in I/O error path
RH-Jira: RHEL-7360
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [4/5] c0b37cc9ff308de300e45d1a10942d11fbabea26 (hreitz/qemu-kvm-c-9-s)
We duplicate the same condition three times here, pull it out to the top
level.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230824155345.109765-5-hreitz@redhat.com>
Reviewed-by: Sam Li <faithilikerun@gmail.com>
(cherry picked from commit d31b50a15dd25a560749b25fc40b6484fd1a57b7)
Downstream conflict with the downstream change in HEAD^^ (conflicting
code is removed).
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/file-posix.c | 18 +++++-------------
1 file changed, 5 insertions(+), 13 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index b9d5e4741b..aa89789737 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -2506,11 +2506,10 @@ static int coroutine_fn raw_co_prw(BlockDriverState *bs, uint64_t offset,
out:
#if defined(CONFIG_BLKZONED)
-{
- BlockZoneWps *wps = bs->wps;
- if (ret == 0) {
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
- bs->bl.zoned != BLK_Z_NONE) {
+ if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
+ bs->bl.zoned != BLK_Z_NONE) {
+ BlockZoneWps *wps = bs->wps;
+ if (ret == 0) {
uint64_t *wp = &wps->wp[offset / bs->bl.zone_size];
if (!BDRV_ZT_IS_CONV(*wp)) {
if (type & QEMU_AIO_ZONE_APPEND) {
@@ -2523,19 +2522,12 @@ out:
*wp = offset + bytes;
}
}
- }
- } else {
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
- bs->bl.zoned != BLK_Z_NONE) {
+ } else {
update_zones_wp(bs, s->fd, 0, 1);
}
- }
- if ((type & (QEMU_AIO_WRITE | QEMU_AIO_ZONE_APPEND)) &&
- bs->bl.zoned != BLK_Z_NONE) {
qemu_co_mutex_unlock(&wps->colock);
}
-}
#endif
return ret;
}
--
2.39.3

57
kvm-fix-clang17-s390x.patch Normal file
View File

@ -0,0 +1,57 @@
From 90733c05b4fe0c55b6427a812ca262d7121fbaa4 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Thu, 19 Oct 2023 02:55:22 -0400
Subject: [PATCH] Fix build s390x build issue with Clang 17
When using Clang 17, s390x build fails due to unaligned variables.
This is fix for the issue.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
host/include/generic/host/atomic128-cas.h | 2 +-
host/include/generic/host/atomic128-ldst.h | 2 +-
include/qemu/int128.h | 1 +
3 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/host/include/generic/host/atomic128-cas.h b/host/include/generic/host/atomic128-cas.h
index 991d3da082..6b40cc2271 100644
--- a/host/include/generic/host/atomic128-cas.h
+++ b/host/include/generic/host/atomic128-cas.h
@@ -28,7 +28,7 @@ atomic16_cmpxchg(Int128 *ptr, Int128 cmp, Int128 new)
static inline Int128 ATTRIBUTE_ATOMIC128_OPT
atomic16_cmpxchg(Int128 *ptr, Int128 cmp, Int128 new)
{
- __int128_t *ptr_align = __builtin_assume_aligned(ptr, 16);
+ Int128Aligned *ptr_align = __builtin_assume_aligned(ptr, 16);
Int128Alias r, c, n;
c.s = cmp;
diff --git a/host/include/generic/host/atomic128-ldst.h b/host/include/generic/host/atomic128-ldst.h
index 80fff0643a..691e6a8531 100644
--- a/host/include/generic/host/atomic128-ldst.h
+++ b/host/include/generic/host/atomic128-ldst.h
@@ -58,7 +58,7 @@ atomic16_read_rw(Int128 *ptr)
static inline void ATTRIBUTE_ATOMIC128_OPT
atomic16_set(Int128 *ptr, Int128 val)
{
- __int128_t *ptr_align = __builtin_assume_aligned(ptr, 16);
+ Int128Aligned *ptr_align = __builtin_assume_aligned(ptr, 16);
__int128_t old;
Int128Alias new;
diff --git a/include/qemu/int128.h b/include/qemu/int128.h
index 73624e8be7..44530d3e10 100644
--- a/include/qemu/int128.h
+++ b/include/qemu/int128.h
@@ -10,6 +10,7 @@
*/
#if defined(CONFIG_INT128) && !defined(CONFIG_TCG_INTERPRETER)
typedef __int128_t Int128;
+typedef __int128_t __attribute__((aligned(16))) Int128Aligned;
static inline Int128 int128_make64(uint64_t a)
{
--
2.39.3

86
kvm-migration-Consolidate-return-path-closing-code.patch Normal file
View File

@ -0,0 +1,86 @@
From 33f0f63da753fff2a3d2bf14542abe0665878e3d Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:20 -0300
Subject: [PATCH 06/13] migration: Consolidate return path closing code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [6/8] d3a4e08e1f1afe92afac12a3404a84d1c23e0f02 (peterx/qemu-kvm)
We'll start calling the await_return_path_close_on_source() function
from other parts of the code, so move all of the related checks and
tracepoints into it.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-7-farosas@suse.de>
(cherry picked from commit d50f5dc075cbb891bfe4a9378600a4871264468a)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 29 ++++++++++++++---------------
1 file changed, 14 insertions(+), 15 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index e2e4a7d8ae..ac4541b971 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -2050,6 +2050,14 @@ static int open_return_path_on_source(MigrationState *ms,
/* Returns 0 if the RP was ok, otherwise there was an error on the RP */
static int await_return_path_close_on_source(MigrationState *ms)
{
+ int ret;
+
+ if (!ms->rp_state.rp_thread_created) {
+ return 0;
+ }
+
+ trace_migration_return_path_end_before();
+
/*
* If this is a normal exit then the destination will send a SHUT
* and the rp_thread will exit, however if there's an error we
@@ -2067,7 +2075,10 @@ static int await_return_path_close_on_source(MigrationState *ms)
qemu_thread_join(&ms->rp_state.rp_thread);
ms->rp_state.rp_thread_created = false;
trace_await_return_path_close_on_source_close();
- return ms->rp_state.error;
+
+ ret = ms->rp_state.error;
+ trace_migration_return_path_end_after(ret);
+ return ret;
}
static inline void
@@ -2363,20 +2374,8 @@ static void migration_completion(MigrationState *s)
goto fail;
}
- /*
- * If rp was opened we must clean up the thread before
- * cleaning everything else up (since if there are no failures
- * it will wait for the destination to send it's status in
- * a SHUT command).
- */
- if (s->rp_state.rp_thread_created) {
- int rp_error;
- trace_migration_return_path_end_before();
- rp_error = await_return_path_close_on_source(s);
- trace_migration_return_path_end_after(rp_error);
- if (rp_error) {
- goto fail;
- }
+ if (await_return_path_close_on_source(s)) {
+ goto fail;
}
if (qemu_file_get_error(s->to_dst_file)) {
--
2.39.3

48
kvm-migration-Fix-possible-race-when-setting-rp_state.er.patch Normal file
View File

@ -0,0 +1,48 @@
From 93aa615c5e8de7422d7162a0731739b2bd82a582 Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:16 -0300
Subject: [PATCH 02/13] migration: Fix possible race when setting
rp_state.error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [2/8] 97305847fc18b6b18cb6f3ff32c82918c2562664 (peterx/qemu-kvm)
We don't need to set the rp_state.error right after a shutdown because
qemu_file_shutdown() always sets the QEMUFile error, so the return
path thread would have seen it and set the rp error itself.
Setting the error outside of the thread is also racy because the
thread could clear it after we set it.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-3-farosas@suse.de>
(cherry picked from commit 28a8347281e24c2e7bba6d3301472eda41d4c096)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/migration/migration.c b/migration/migration.c
index 6a7122694a..b92c6ae436 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -2063,7 +2063,6 @@ static int await_return_path_close_on_source(MigrationState *ms)
* waiting for the destination.
*/
qemu_file_shutdown(ms->rp_state.from_dst_file);
- mark_source_rp_bad(ms);
}
trace_await_return_path_close_on_source_joining();
qemu_thread_join(&ms->rp_state.rp_thread);
--
2.39.3

82
kvm-migration-Fix-possible-race-when-shutting-down-to_ds.patch Normal file
View File

@ -0,0 +1,82 @@
From 52e2456a41678f29a67e4b7cd4d7ee20e5298bc0 Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:18 -0300
Subject: [PATCH 04/13] migration: Fix possible race when shutting down
to_dst_file
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [4/8] 99b9ce27c4477b9630cedfb02a534f8c490ab1eb (peterx/qemu-kvm)
It's not safe to call qemu_file_shutdown() on the to_dst_file without
first checking for the file's presence under the lock. The cleanup of
this file happens at postcopy_pause() and migrate_fd_cleanup() which
are not necessarily running in the same thread as migrate_fd_cancel().
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-5-farosas@suse.de>
(cherry picked from commit 7478fb0df914f0a5ab551ff74b1df62dd250500e)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 517b3e04d2..169e6bdce8 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1246,7 +1246,7 @@ static void migrate_fd_error(MigrationState *s, const Error *error)
static void migrate_fd_cancel(MigrationState *s)
{
int old_state ;
- QEMUFile *f = migrate_get_current()->to_dst_file;
+
trace_migrate_fd_cancel();
WITH_QEMU_LOCK_GUARD(&s->qemu_file_lock) {
@@ -1272,11 +1272,13 @@ static void migrate_fd_cancel(MigrationState *s)
* If we're unlucky the migration code might be stuck somewhere in a
* send/write while the network has failed and is waiting to timeout;
* if we've got shutdown(2) available then we can force it to quit.
- * The outgoing qemu file gets closed in migrate_fd_cleanup that is
- * called in a bh, so there is no race against this cancel.
*/
- if (s->state == MIGRATION_STATUS_CANCELLING && f) {
- qemu_file_shutdown(f);
+ if (s->state == MIGRATION_STATUS_CANCELLING) {
+ WITH_QEMU_LOCK_GUARD(&s->qemu_file_lock) {
+ if (s->to_dst_file) {
+ qemu_file_shutdown(s->to_dst_file);
+ }
+ }
}
if (s->state == MIGRATION_STATUS_CANCELLING && s->block_inactive) {
Error *local_err = NULL;
@@ -1520,12 +1522,14 @@ void qmp_migrate_pause(Error **errp)
{
MigrationState *ms = migrate_get_current();
MigrationIncomingState *mis = migration_incoming_get_current();
- int ret;
+ int ret = 0;
if (ms->state == MIGRATION_STATUS_POSTCOPY_ACTIVE) {
/* Source side, during postcopy */
qemu_mutex_lock(&ms->qemu_file_lock);
- ret = qemu_file_shutdown(ms->to_dst_file);
+ if (ms->to_dst_file) {
+ ret = qemu_file_shutdown(ms->to_dst_file);
+ }
qemu_mutex_unlock(&ms->qemu_file_lock);
if (ret) {
error_setg(errp, "Failed to pause source migration");
--
2.39.3

74
kvm-migration-Fix-possible-races-when-shutting-down-the-.patch Normal file
View File

@ -0,0 +1,74 @@
From 1cc26aac30b883244e300d872fa3a19f39afbb66 Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:17 -0300
Subject: [PATCH 03/13] migration: Fix possible races when shutting down the
return path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [3/8] d734387cf07e956a1f1e817c200e5b36cf0a00b1 (peterx/qemu-kvm)
We cannot call qemu_file_shutdown() on the return path file without
taking the file lock. The return path thread could be running it's
cleanup code and have just cleared the from_dst_file pointer.
Checking ms->to_dst_file for errors could also race with
migrate_fd_cleanup() which clears the to_dst_file pointer.
Protect both accesses by taking the file lock.
This was caught by inspection, it should be rare, but the next patches
will start calling this code from other places, so let's do the
correct thing.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-4-farosas@suse.de>
(cherry picked from commit 639decf529793fc544c8055b82be8abe77fa48fa)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index b92c6ae436..517b3e04d2 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -2053,17 +2053,18 @@ static int open_return_path_on_source(MigrationState *ms,
static int await_return_path_close_on_source(MigrationState *ms)
{
/*
- * If this is a normal exit then the destination will send a SHUT and the
- * rp_thread will exit, however if there's an error we need to cause
- * it to exit.
+ * If this is a normal exit then the destination will send a SHUT
+ * and the rp_thread will exit, however if there's an error we
+ * need to cause it to exit. shutdown(2), if we have it, will
+ * cause it to unblock if it's stuck waiting for the destination.
*/
- if (qemu_file_get_error(ms->to_dst_file) && ms->rp_state.from_dst_file) {
- /*
- * shutdown(2), if we have it, will cause it to unblock if it's stuck
- * waiting for the destination.
- */
- qemu_file_shutdown(ms->rp_state.from_dst_file);
+ WITH_QEMU_LOCK_GUARD(&ms->qemu_file_lock) {
+ if (ms->to_dst_file && ms->rp_state.from_dst_file &&
+ qemu_file_get_error(ms->to_dst_file)) {
+ qemu_file_shutdown(ms->rp_state.from_dst_file);
+ }
}
+
trace_await_return_path_close_on_source_joining();
qemu_thread_join(&ms->rp_state.rp_thread);
ms->rp_state.rp_thread_created = false;
--
2.39.3

172
kvm-migration-Fix-race-that-dest-preempt-thread-close-to.patch Normal file
View File

@ -0,0 +1,172 @@
From c98eb88ce49927f612753b53c890135fde14c3a4 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Mon, 18 Sep 2023 14:28:15 -0300
Subject: [PATCH 01/13] migration: Fix race that dest preempt thread close too
early
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [1/8] a1197ca99878d5d8bbb271c6941aa36d91c66122 (peterx/qemu-kvm)
We hit intermit CI issue on failing at migration-test over the unit test
preempt/plain:
qemu-system-x86_64: Unable to read from socket: Connection reset by peer
Memory content inconsistency at 5b43000 first_byte = bd last_byte = bc current = 4f hit_edge = 1
**
ERROR:../tests/qtest/migration-test.c:300:check_guests_ram: assertion failed: (bad == 0)
(test program exited with status code -6)
Fabiano debugged into it and found that the preempt thread can quit even
without receiving all the pages, which can cause guest not receiving all
the pages and corrupt the guest memory.
To make sure preempt thread finished receiving all the pages, we can rely
on the page_requested_count being zero because preempt channel will only
receive requested page faults. Note, not all the faulted pages are required
to be sent via the preempt channel/thread; imagine the case when a
requested page is just queued into the background main channel for
migration, the src qemu will just still send it via the background channel.
Here instead of spinning over reading the count, we add a condvar so the
main thread can wait on it if that unusual case happened, without burning
the cpu for no good reason, even if the duration is short; so even if we
spin in this rare case is probably fine. It's just better to not do so.
The condvar is only used when that special case is triggered. Some memory
ordering trick is needed to guarantee it from happening (against the
preempt thread status field), so the main thread will always get a kick
when that triggers correctly.
Closes: https://gitlab.com/qemu-project/qemu/-/issues/1886
Debugged-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-2-farosas@suse.de>
(cherry picked from commit cf02f29e1e3843784630d04783e372fa541a77e5)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 3 ++-
migration/migration.h | 13 ++++++++++++-
migration/postcopy-ram.c | 38 +++++++++++++++++++++++++++++++++++++-
3 files changed, 51 insertions(+), 3 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 5528acb65e..6a7122694a 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -153,6 +153,7 @@ void migration_object_init(void)
qemu_sem_init(&current_incoming->postcopy_qemufile_dst_done, 0);
qemu_mutex_init(&current_incoming->page_request_mutex);
+ qemu_cond_init(&current_incoming->page_request_cond);
current_incoming->page_requested = g_tree_new(page_request_addr_cmp);
migration_object_check(current_migration, &error_fatal);
@@ -367,7 +368,7 @@ int migrate_send_rp_req_pages(MigrationIncomingState *mis,
* things like g_tree_lookup() will return TRUE (1) when found.
*/
g_tree_insert(mis->page_requested, aligned, (gpointer)1);
- mis->page_requested_count++;
+ qatomic_inc(&mis->page_requested_count);
trace_postcopy_page_req_add(aligned, mis->page_requested_count);
}
}
diff --git a/migration/migration.h b/migration/migration.h
index 6eea18db36..9a30216895 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -196,7 +196,10 @@ struct MigrationIncomingState {
/* A tree of pages that we requested to the source VM */
GTree *page_requested;
- /* For debugging purpose only, but would be nice to keep */
+ /*
+ * For postcopy only, count the number of requested page faults that
+ * still haven't been resolved.
+ */
int page_requested_count;
/*
* The mutex helps to maintain the requested pages that we sent to the
@@ -210,6 +213,14 @@ struct MigrationIncomingState {
* contains valid information.
*/
QemuMutex page_request_mutex;
+ /*
+ * If postcopy preempt is enabled, there is a chance that the main
+ * thread finished loading its data before the preempt channel has
+ * finished loading the urgent pages. If that happens, the two threads
+ * will use this condvar to synchronize, so the main thread will always
+ * wait until all pages received.
+ */
+ QemuCond page_request_cond;
/*
* Number of devices that have yet to approve switchover. When this reaches
diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c
index 29aea9456d..5408e028c6 100644
--- a/migration/postcopy-ram.c
+++ b/migration/postcopy-ram.c
@@ -599,6 +599,30 @@ int postcopy_ram_incoming_cleanup(MigrationIncomingState *mis)
if (mis->preempt_thread_status == PREEMPT_THREAD_CREATED) {
/* Notify the fast load thread to quit */
mis->preempt_thread_status = PREEMPT_THREAD_QUIT;
+ /*
+ * Update preempt_thread_status before reading count. Note: mutex
+ * lock only provide ACQUIRE semantic, and it doesn't stops this
+ * write to be reordered after reading the count.
+ */
+ smp_mb();
+ /*
+ * It's possible that the preempt thread is still handling the last
+ * pages to arrive which were requested by guest page faults.
+ * Making sure nothing is left behind by waiting on the condvar if
+ * that unlikely case happened.
+ */
+ WITH_QEMU_LOCK_GUARD(&mis->page_request_mutex) {
+ if (qatomic_read(&mis->page_requested_count)) {
+ /*
+ * It is guaranteed to receive a signal later, because the
+ * count>0 now, so it's destined to be decreased to zero
+ * very soon by the preempt thread.
+ */
+ qemu_cond_wait(&mis->page_request_cond,
+ &mis->page_request_mutex);
+ }
+ }
+ /* Notify the fast load thread to quit */
if (mis->postcopy_qemufile_dst) {
qemu_file_shutdown(mis->postcopy_qemufile_dst);
}
@@ -1277,8 +1301,20 @@ static int qemu_ufd_copy_ioctl(MigrationIncomingState *mis, void *host_addr,
*/
if (g_tree_lookup(mis->page_requested, host_addr)) {
g_tree_remove(mis->page_requested, host_addr);
- mis->page_requested_count--;
+ int left_pages = qatomic_dec_fetch(&mis->page_requested_count);
+
trace_postcopy_page_req_del(host_addr, mis->page_requested_count);
+ /* Order the update of count and read of preempt status */
+ smp_mb();
+ if (mis->preempt_thread_status == PREEMPT_THREAD_QUIT &&
+ left_pages == 0) {
+ /*
+ * This probably means the main thread is waiting for us.
+ * Notify that we've finished receiving the last requested
+ * page.
+ */
+ qemu_cond_signal(&mis->page_request_cond);
+ }
}
qemu_mutex_unlock(&mis->page_request_mutex);
mark_postcopy_blocktime_end((uintptr_t)host_addr);
--
2.39.3

76
kvm-migration-Move-return-path-cleanup-to-main-migration.patch Normal file
View File

@ -0,0 +1,76 @@
From 117f325d4983bb7e40a50e51c2fe056938d5add4 Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:22 -0300
Subject: [PATCH 08/13] migration: Move return path cleanup to main migration
thread
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [8/8] 3f085c4ac6c0835d35346bf0cad778c17a248daf (peterx/qemu-kvm)
Now that the return path thread is allowed to finish during a paused
migration, we can move the cleanup of the QEMUFiles to the main
migration thread.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-9-farosas@suse.de>
(cherry picked from commit 36e9aab3c569d4c9ad780473596e18479838d1aa)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/migration/migration.c b/migration/migration.c
index a0782c64a1..7a4c8beb5d 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -98,6 +98,7 @@ static int migration_maybe_pause(MigrationState *s,
int *current_active_state,
int new_state);
static void migrate_fd_cancel(MigrationState *s);
+static int await_return_path_close_on_source(MigrationState *s);
static bool migration_needs_multiple_sockets(void)
{
@@ -1178,6 +1179,12 @@ static void migrate_fd_cleanup(MigrationState *s)
qemu_fclose(tmp);
}
+ /*
+ * We already cleaned up to_dst_file, so errors from the return
+ * path might be due to that, ignore them.
+ */
+ await_return_path_close_on_source(s);
+
assert(!migration_is_active(s));
if (s->state == MIGRATION_STATUS_CANCELLING) {
@@ -1986,7 +1993,6 @@ out:
}
trace_source_return_path_thread_end();
- migration_release_dst_files(ms);
rcu_unregister_thread();
return NULL;
}
@@ -2040,6 +2046,9 @@ static int await_return_path_close_on_source(MigrationState *ms)
ret = ms->rp_state.error;
ms->rp_state.error = false;
+
+ migration_release_dst_files(ms);
+
trace_migration_return_path_end_after(ret);
return ret;
}
--
2.39.3

49
kvm-migration-Remove-redundant-cleanup-of-postcopy_qemuf.patch Normal file
View File

@ -0,0 +1,49 @@
From be84138fc49f3e6998d6af8fdfa9cc1054ae1549 Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:19 -0300
Subject: [PATCH 05/13] migration: Remove redundant cleanup of
postcopy_qemufile_src
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [5/8] 2a63a021e923e6460d57f1c0850651f4ddcd5b4f (peterx/qemu-kvm)
This file is owned by the return path thread which is already doing
cleanup.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-6-farosas@suse.de>
(cherry picked from commit b3b101157d4651f12e6b3361af2de6bace7f9b4a)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 6 ------
1 file changed, 6 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index 169e6bdce8..e2e4a7d8ae 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1178,12 +1178,6 @@ static void migrate_fd_cleanup(MigrationState *s)
qemu_fclose(tmp);
}
- if (s->postcopy_qemufile_src) {
- migration_ioc_unregister_yank_from_file(s->postcopy_qemufile_src);
- qemu_fclose(s->postcopy_qemufile_src);
- s->postcopy_qemufile_src = NULL;
- }
-
assert(!migration_is_active(s));
if (s->state == MIGRATION_STATUS_CANCELLING) {
--
2.39.3

251
kvm-migration-Replace-the-return-path-retry-logic.patch Normal file
View File

@ -0,0 +1,251 @@
From 74093a6ac7b5290965a2869512b998eac018a630 Mon Sep 17 00:00:00 2001
From: Fabiano Rosas <farosas@suse.de>
Date: Mon, 18 Sep 2023 14:28:21 -0300
Subject: [PATCH 07/13] migration: Replace the return path retry logic
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Peter Xu <peterx@redhat.com>
RH-MergeRequest: 203: migration: Fix race that dest preempt thread close too early
RH-Jira: RHEL-11219
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Leonardo Brás <leobras@redhat.com>
RH-Commit: [7/8] 6b9d9b955f95ff14b47491865e7e45bfdcfc8a97 (peterx/qemu-kvm)
Replace the return path retry logic with finishing and restarting the
thread. This fixes a race when resuming the migration that leads to a
segfault.
Currently when doing postcopy we consider that an IO error on the
return path file could be due to a network intermittency. We then keep
the thread alive but have it do cleanup of the 'from_dst_file' and
wait on the 'postcopy_pause_rp' semaphore. When the user issues a
migrate resume, a new return path is opened and the thread is allowed
to continue.
There's a race condition in the above mechanism. It is possible for
the new return path file to be setup *before* the cleanup code in the
return path thread has had a chance to run, leading to the *new* file
being closed and the pointer set to NULL. When the thread is released
after the resume, it tries to dereference 'from_dst_file' and crashes:
Thread 7 "return path" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffd1dbf700 (LWP 9611)]
0x00005555560e4893 in qemu_file_get_error_obj (f=0x0, errp=0x0) at ../migration/qemu-file.c:154
154 return f->last_error;
(gdb) bt
#0 0x00005555560e4893 in qemu_file_get_error_obj (f=0x0, errp=0x0) at ../migration/qemu-file.c:154
#1 0x00005555560e4983 in qemu_file_get_error (f=0x0) at ../migration/qemu-file.c:206
#2 0x0000555555b9a1df in source_return_path_thread (opaque=0x555556e06000) at ../migration/migration.c:1876
#3 0x000055555602e14f in qemu_thread_start (args=0x55555782e780) at ../util/qemu-thread-posix.c:541
#4 0x00007ffff38d76ea in start_thread (arg=0x7fffd1dbf700) at pthread_create.c:477
#5 0x00007ffff35efa6f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
Here's the race (important bit is open_return_path happening before
migration_release_dst_files):
migration | qmp | return path
--------------------------+-----------------------------+---------------------------------
qmp_migrate_pause()
shutdown(ms->to_dst_file)
f->last_error = -EIO
migrate_detect_error()
postcopy_pause()
set_state(PAUSED)
wait(postcopy_pause_sem)
qmp_migrate(resume)
migrate_fd_connect()
resume = state == PAUSED
open_return_path <-- TOO SOON!
set_state(RECOVER)
post(postcopy_pause_sem)
(incoming closes to_src_file)
res = qemu_file_get_error(rp)
migration_release_dst_files()
ms->rp_state.from_dst_file = NULL
post(postcopy_pause_rp_sem)
postcopy_pause_return_path_thread()
wait(postcopy_pause_rp_sem)
rp = ms->rp_state.from_dst_file
goto retry
qemu_file_get_error(rp)
SIGSEGV
-------------------------------------------------------------------------------------------
We can keep the retry logic without having the thread alive and
waiting. The only piece of data used by it is the 'from_dst_file' and
it is only allowed to proceed after a migrate resume is issued and the
semaphore released at migrate_fd_connect().
Move the retry logic to outside the thread by waiting for the thread
to finish before pausing the migration.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230918172822.19052-8-farosas@suse.de>
(cherry picked from commit ef796ee93b313ed2f0b427ef30320417387d2ad5)
Signed-off-by: Peter Xu <peterx@redhat.com>
---
migration/migration.c | 60 ++++++++-----------------------------------
migration/migration.h | 1 -
2 files changed, 11 insertions(+), 50 deletions(-)
diff --git a/migration/migration.c b/migration/migration.c
index ac4541b971..a0782c64a1 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1776,18 +1776,6 @@ static void migrate_handle_rp_req_pages(MigrationState *ms, const char* rbname,
}
}
-/* Return true to retry, false to quit */
-static bool postcopy_pause_return_path_thread(MigrationState *s)
-{
- trace_postcopy_pause_return_path();
-
- qemu_sem_wait(&s->postcopy_pause_rp_sem);
-
- trace_postcopy_pause_return_path_continued();
-
- return true;
-}
-
static int migrate_handle_rp_recv_bitmap(MigrationState *s, char *block_name)
{
RAMBlock *block = qemu_ram_block_by_name(block_name);
@@ -1871,7 +1859,6 @@ static void *source_return_path_thread(void *opaque)
trace_source_return_path_thread_entry();
rcu_register_thread();
-retry:
while (!ms->rp_state.error && !qemu_file_get_error(rp) &&
migration_is_setup_or_active(ms->state)) {
trace_source_return_path_thread_loop_top();
@@ -1993,26 +1980,7 @@ retry:
}
out:
- res = qemu_file_get_error(rp);
- if (res) {
- if (res && migration_in_postcopy()) {
- /*
- * Maybe there is something we can do: it looks like a
- * network down issue, and we pause for a recovery.
- */
- migration_release_dst_files(ms);
- rp = NULL;
- if (postcopy_pause_return_path_thread(ms)) {
- /*
- * Reload rp, reset the rest. Referencing it is safe since
- * it's reset only by us above, or when migration completes
- */
- rp = ms->rp_state.from_dst_file;
- ms->rp_state.error = false;
- goto retry;
- }
- }
-
+ if (qemu_file_get_error(rp)) {
trace_source_return_path_thread_bad_end();
mark_source_rp_bad(ms);
}
@@ -2023,8 +1991,7 @@ out:
return NULL;
}
-static int open_return_path_on_source(MigrationState *ms,
- bool create_thread)
+static int open_return_path_on_source(MigrationState *ms)
{
ms->rp_state.from_dst_file = qemu_file_get_return_path(ms->to_dst_file);
if (!ms->rp_state.from_dst_file) {
@@ -2033,11 +2000,6 @@ static int open_return_path_on_source(MigrationState *ms,
trace_open_return_path_on_source();
- if (!create_thread) {
- /* We're done */
- return 0;
- }
-
qemu_thread_create(&ms->rp_state.rp_thread, "return path",
source_return_path_thread, ms, QEMU_THREAD_JOINABLE);
ms->rp_state.rp_thread_created = true;
@@ -2077,6 +2039,7 @@ static int await_return_path_close_on_source(MigrationState *ms)
trace_await_return_path_close_on_source_close();
ret = ms->rp_state.error;
+ ms->rp_state.error = false;
trace_migration_return_path_end_after(ret);
return ret;
}
@@ -2552,6 +2515,13 @@ static MigThrError postcopy_pause(MigrationState *s)
qemu_file_shutdown(file);
qemu_fclose(file);
+ /*
+ * We're already pausing, so ignore any errors on the return
+ * path and just wait for the thread to finish. It will be
+ * re-created when we resume.
+ */
+ await_return_path_close_on_source(s);
+
migrate_set_state(&s->state, s->state,
MIGRATION_STATUS_POSTCOPY_PAUSED);
@@ -2569,12 +2539,6 @@ static MigThrError postcopy_pause(MigrationState *s)
if (s->state == MIGRATION_STATUS_POSTCOPY_RECOVER) {
/* Woken up by a recover procedure. Give it a shot */
- /*
- * Firstly, let's wake up the return path now, with a new
- * return path channel.
- */
- qemu_sem_post(&s->postcopy_pause_rp_sem);
-
/* Do the resume logic */
if (postcopy_do_resume(s) == 0) {
/* Let's continue! */
@@ -3264,7 +3228,7 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
* QEMU uses the return path.
*/
if (migrate_postcopy_ram() || migrate_return_path()) {
- if (open_return_path_on_source(s, !resume)) {
+ if (open_return_path_on_source(s)) {
error_setg(&local_err, "Unable to open return-path for postcopy");
migrate_set_state(&s->state, s->state, MIGRATION_STATUS_FAILED);
migrate_set_error(s, local_err);
@@ -3328,7 +3292,6 @@ static void migration_instance_finalize(Object *obj)
qemu_sem_destroy(&ms->rate_limit_sem);
qemu_sem_destroy(&ms->pause_sem);
qemu_sem_destroy(&ms->postcopy_pause_sem);
- qemu_sem_destroy(&ms->postcopy_pause_rp_sem);
qemu_sem_destroy(&ms->rp_state.rp_sem);
qemu_sem_destroy(&ms->rp_state.rp_pong_acks);
qemu_sem_destroy(&ms->postcopy_qemufile_src_sem);
@@ -3348,7 +3311,6 @@ static void migration_instance_init(Object *obj)
migrate_params_init(&ms->parameters);
qemu_sem_init(&ms->postcopy_pause_sem, 0);
- qemu_sem_init(&ms->postcopy_pause_rp_sem, 0);
qemu_sem_init(&ms->rp_state.rp_sem, 0);
qemu_sem_init(&ms->rp_state.rp_pong_acks, 0);
qemu_sem_init(&ms->rate_limit_sem, 0);
diff --git a/migration/migration.h b/migration/migration.h
index 9a30216895..1034d617bf 100644
--- a/migration/migration.h
+++ b/migration/migration.h
@@ -393,7 +393,6 @@ struct MigrationState {
/* Needed by postcopy-pause state */
QemuSemaphore postcopy_pause_sem;
- QemuSemaphore postcopy_pause_rp_sem;
/*
* Whether we abort the migration if decompression errors are
* detected at the destination. It is left at false for qemu
--
2.39.3

201
kvm-tests-file-io-error-New-test.patch Normal file
View File

@ -0,0 +1,201 @@
From dbf091dd32daae0999fb62b74f7c834b445f6bf1 Mon Sep 17 00:00:00 2001
From: Hanna Czenczek <hreitz@redhat.com>
Date: Thu, 24 Aug 2023 17:53:44 +0200
Subject: [PATCH 13/13] tests/file-io-error: New test
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 202: file-posix: Fix zone update in I/O error path
RH-Jira: RHEL-7360
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [5/5] a07662c948f10bea7972a7d525b5941264651a4f (hreitz/qemu-kvm-c-9-s)
This is a regression test for
https://bugzilla.redhat.com/show_bug.cgi?id=2234374.
All this test needs to do is trigger an I/O error inside of file-posix
(specifically raw_co_prw()). One reliable way to do this without
requiring special privileges is to use a FUSE export, which allows us to
inject any error that we want, e.g. via blkdebug.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230824155345.109765-6-hreitz@redhat.com>
[hreitz: Fixed test to be skipped when there is no FUSE support, to
suppress fusermount's allow_other warning, and to be skipped
with $IMGOPTSSYNTAX enabled]
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
(cherry picked from commit 380448464dd89291cf7fd7434be6c225482a334d)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
tests/qemu-iotests/tests/file-io-error | 119 +++++++++++++++++++++
tests/qemu-iotests/tests/file-io-error.out | 33 ++++++
2 files changed, 152 insertions(+)
create mode 100755 tests/qemu-iotests/tests/file-io-error
create mode 100644 tests/qemu-iotests/tests/file-io-error.out
diff --git a/tests/qemu-iotests/tests/file-io-error b/tests/qemu-iotests/tests/file-io-error
new file mode 100755
index 0000000000..88ee5f670c
--- /dev/null
+++ b/tests/qemu-iotests/tests/file-io-error
@@ -0,0 +1,119 @@
+#!/usr/bin/env bash
+# group: rw
+#
+# Produce an I/O error in file-posix, and hope that it is not catastrophic.
+# Regression test for: https://bugzilla.redhat.com/show_bug.cgi?id=2234374
+#
+# Copyright (C) 2023 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+seq=$(basename "$0")
+echo "QA output created by $seq"
+
+status=1 # failure is the default!
+
+_cleanup()
+{
+ _cleanup_qemu
+ rm -f "$TEST_DIR/fuse-export"
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ../common.rc
+. ../common.filter
+. ../common.qemu
+
+# Format-agnostic (we do not use any), but we do test the file protocol
+_supported_proto file
+_require_drivers blkdebug null-co
+
+if [ "$IMGOPTSSYNTAX" = "true" ]; then
+ # We need `$QEMU_IO -f file` to work; IMGOPTSSYNTAX uses --image-opts,
+ # breaking -f.
+ _unsupported_fmt $IMGFMT
+fi
+
+# This is a regression test of a bug in which flie-posix would access zone
+# information in case of an I/O error even when there is no zone information,
+# resulting in a division by zero.
+# To reproduce the problem, we need to trigger an I/O error inside of
+# file-posix, which can be done (rootless) by providing a FUSE export that
+# presents only errors when accessed.
+
+_launch_qemu
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'qmp_capabilities'}" \
+ 'return'
+
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'blockdev-add',
+ 'arguments': {
+ 'driver': 'blkdebug',
+ 'node-name': 'node0',
+ 'inject-error': [{'event': 'none'}],
+ 'image': {
+ 'driver': 'null-co'
+ }
+ }}" \
+ 'return'
+
+# FUSE mountpoint must exist and be a regular file
+touch "$TEST_DIR/fuse-export"
+
+# The grep -v to filter fusermount's (benign) error when /etc/fuse.conf does
+# not contain user_allow_other and the subsequent check for missing FUSE support
+# have both been taken from iotest 308.
+output=$(_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'block-export-add',
+ 'arguments': {
+ 'id': 'exp0',
+ 'type': 'fuse',
+ 'node-name': 'node0',
+ 'mountpoint': '$TEST_DIR/fuse-export',
+ 'writable': true
+ }}" \
+ 'return' \
+ | grep -v 'option allow_other only allowed if')
+
+if echo "$output" | grep -q "Parameter 'type' does not accept value 'fuse'"; then
+ _notrun 'No FUSE support'
+fi
+echo "$output"
+
+echo
+# This should fail, but gracefully, i.e. just print an I/O error, not crash.
+$QEMU_IO -f file -c 'write 0 64M' "$TEST_DIR/fuse-export" | _filter_qemu_io
+echo
+
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'block-export-del',
+ 'arguments': {'id': 'exp0'}}" \
+ 'return'
+
+_send_qemu_cmd $QEMU_HANDLE \
+ '' \
+ 'BLOCK_EXPORT_DELETED'
+
+_send_qemu_cmd $QEMU_HANDLE \
+ "{'execute': 'blockdev-del',
+ 'arguments': {'node-name': 'node0'}}" \
+ 'return'
+
+# success, all done
+echo "*** done"
+rm -f $seq.full
+status=0
diff --git a/tests/qemu-iotests/tests/file-io-error.out b/tests/qemu-iotests/tests/file-io-error.out
new file mode 100644
index 0000000000..0f46455a94
--- /dev/null
+++ b/tests/qemu-iotests/tests/file-io-error.out
@@ -0,0 +1,33 @@
+QA output created by file-io-error
+{'execute': 'qmp_capabilities'}
+{"return": {}}
+{'execute': 'blockdev-add',
+ 'arguments': {
+ 'driver': 'blkdebug',
+ 'node-name': 'node0',
+ 'inject-error': [{'event': 'none'}],
+ 'image': {
+ 'driver': 'null-co'
+ }
+ }}
+{"return": {}}
+{'execute': 'block-export-add',
+ 'arguments': {
+ 'id': 'exp0',
+ 'type': 'fuse',
+ 'node-name': 'node0',
+ 'mountpoint': 'TEST_DIR/fuse-export',
+ 'writable': true
+ }}
+{"return": {}}
+
+write failed: Input/output error
+
+{'execute': 'block-export-del',
+ 'arguments': {'id': 'exp0'}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_EXPORT_DELETED", "data": {"id": "exp0"}}
+{'execute': 'blockdev-del',
+ 'arguments': {'node-name': 'node0'}}
+{"return": {}}
+*** done
--
2.39.3

151
kvm-virtio-Drop-out-of-coroutine-context-in-virtio_load.patch Normal file
View File

@ -0,0 +1,151 @@
From be5adda1bfc78fea3655ee7a2e11048cfcd9247c Mon Sep 17 00:00:00 2001
From: Kevin Wolf <kwolf@redhat.com>
Date: Tue, 5 Sep 2023 16:50:02 +0200
Subject: [PATCH] virtio: Drop out of coroutine context in virtio_load()
RH-Author: Kevin Wolf <kwolf@redhat.com>
RH-MergeRequest: 201: virtio: Drop out of coroutine context in virtio_load()
RH-Jira: RHEL-832
RH-Acked-by: Hanna Czenczek <hreitz@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Commit: [1/1] 64ca84cf9bd82eb9e22a1840909e99958ceb51a6 (kmwolf/centos-qemu-kvm)
virtio_load() as a whole should run in coroutine context because it
reads from the migration stream and we don't want this to block.
However, it calls virtio_set_features_nocheck() and devices don't
expect their .set_features callback to run in a coroutine and therefore
call functions that may not be called in coroutine context. To fix this,
drop out of coroutine context for calling virtio_set_features_nocheck().
Without this fix, the following crash was reported:
#0 __pthread_kill_implementation (threadid=<optimized out>, signo=signo@entry=6, no_tid=no_tid@entry=0) at pthread_kill.c:44
#1 0x00007efc738c05d3 in __pthread_kill_internal (signo=6, threadid=<optimized out>) at pthread_kill.c:78
#2 0x00007efc73873d26 in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
#3 0x00007efc738477f3 in __GI_abort () at abort.c:79
#4 0x00007efc7384771b in __assert_fail_base (fmt=0x7efc739dbcb8 "", assertion=assertion@entry=0x560aebfbf5cf "!qemu_in_coroutine()",
file=file@entry=0x560aebfcd2d4 "../block/graph-lock.c", line=line@entry=275, function=function@entry=0x560aebfcd34d "void bdrv_graph_rdlock_main_loop(void)") at assert.c:92
#5 0x00007efc7386ccc6 in __assert_fail (assertion=0x560aebfbf5cf "!qemu_in_coroutine()", file=0x560aebfcd2d4 "../block/graph-lock.c", line=275,
function=0x560aebfcd34d "void bdrv_graph_rdlock_main_loop(void)") at assert.c:101
#6 0x0000560aebcd8dd6 in bdrv_register_buf ()
#7 0x0000560aeb97ed97 in ram_block_added.llvm ()
#8 0x0000560aebb8303f in ram_block_add.llvm ()
#9 0x0000560aebb834fa in qemu_ram_alloc_internal.llvm ()
#10 0x0000560aebb2ac98 in vfio_region_mmap ()
#11 0x0000560aebb3ea0f in vfio_bars_register ()
#12 0x0000560aebb3c628 in vfio_realize ()
#13 0x0000560aeb90f0c2 in pci_qdev_realize ()
#14 0x0000560aebc40305 in device_set_realized ()
#15 0x0000560aebc48e07 in property_set_bool.llvm ()
#16 0x0000560aebc46582 in object_property_set ()
#17 0x0000560aebc4cd58 in object_property_set_qobject ()
#18 0x0000560aebc46ba7 in object_property_set_bool ()
#19 0x0000560aeb98b3ca in qdev_device_add_from_qdict ()
#20 0x0000560aebb1fbaf in virtio_net_set_features ()
#21 0x0000560aebb46b51 in virtio_set_features_nocheck ()
#22 0x0000560aebb47107 in virtio_load ()
#23 0x0000560aeb9ae7ce in vmstate_load_state ()
#24 0x0000560aeb9d2ee9 in qemu_loadvm_state_main ()
#25 0x0000560aeb9d45e1 in qemu_loadvm_state ()
#26 0x0000560aeb9bc32c in process_incoming_migration_co.llvm ()
#27 0x0000560aebeace56 in coroutine_trampoline.llvm ()
Cc: qemu-stable@nongnu.org
Buglink: https://issues.redhat.com/browse/RHEL-832
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20230905145002.46391-3-kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 92e2e6a867334a990f8d29f07ca34e3162fdd6ec)
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
hw/virtio/virtio.c | 45 ++++++++++++++++++++++++++++++++++++++++-----
1 file changed, 40 insertions(+), 5 deletions(-)
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index 309038fd46..969c25f4cf 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -2825,8 +2825,9 @@ static int virtio_device_put(QEMUFile *f, void *opaque, size_t size,
}
/* A wrapper for use as a VMState .get function */
-static int virtio_device_get(QEMUFile *f, void *opaque, size_t size,
- const VMStateField *field)
+static int coroutine_mixed_fn
+virtio_device_get(QEMUFile *f, void *opaque, size_t size,
+ const VMStateField *field)
{
VirtIODevice *vdev = VIRTIO_DEVICE(opaque);
DeviceClass *dc = DEVICE_CLASS(VIRTIO_DEVICE_GET_CLASS(vdev));
@@ -2853,6 +2854,39 @@ static int virtio_set_features_nocheck(VirtIODevice *vdev, uint64_t val)
return bad ? -1 : 0;
}
+typedef struct VirtioSetFeaturesNocheckData {
+ Coroutine *co;
+ VirtIODevice *vdev;
+ uint64_t val;
+ int ret;
+} VirtioSetFeaturesNocheckData;
+
+static void virtio_set_features_nocheck_bh(void *opaque)
+{
+ VirtioSetFeaturesNocheckData *data = opaque;
+
+ data->ret = virtio_set_features_nocheck(data->vdev, data->val);
+ aio_co_wake(data->co);
+}
+
+static int coroutine_mixed_fn
+virtio_set_features_nocheck_maybe_co(VirtIODevice *vdev, uint64_t val)
+{
+ if (qemu_in_coroutine()) {
+ VirtioSetFeaturesNocheckData data = {
+ .co = qemu_coroutine_self(),
+ .vdev = vdev,
+ .val = val,
+ };
+ aio_bh_schedule_oneshot(qemu_get_current_aio_context(),
+ virtio_set_features_nocheck_bh, &data);
+ qemu_coroutine_yield();
+ return data.ret;
+ } else {
+ return virtio_set_features_nocheck(vdev, val);
+ }
+}
+
int virtio_set_features(VirtIODevice *vdev, uint64_t val)
{
int ret;
@@ -2906,7 +2940,8 @@ size_t virtio_get_config_size(const VirtIOConfigSizeParams *params,
return config_size;
}
-int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
+int coroutine_mixed_fn
+virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
{
int i, ret;
int32_t config_len;
@@ -3023,14 +3058,14 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f, int version_id)
* host_features.
*/
uint64_t features64 = vdev->guest_features;
- if (virtio_set_features_nocheck(vdev, features64) < 0) {
+ if (virtio_set_features_nocheck_maybe_co(vdev, features64) < 0) {
error_report("Features 0x%" PRIx64 " unsupported. "
"Allowed features: 0x%" PRIx64,
features64, vdev->host_features);
return -1;
}
} else {
- if (virtio_set_features_nocheck(vdev, features) < 0) {
+ if (virtio_set_features_nocheck_maybe_co(vdev, features) < 0) {
error_report("Features 0x%x unsupported. "
"Allowed features: 0x%" PRIx64,
features, vdev->host_features);
--
2.39.3

98
qemu.spec
View File

@ -149,7 +149,7 @@ Obsoletes: %{name}-block-ssh <= %{epoch}:%{version} \
Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm
Version: 8.1.0
Release: 0%{?rcrel}%{?dist}%{?cc_suffix}
Release: 3%{?rcrel}%{?dist}%{?cc_suffix}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
# Epoch 15 used for RHEL 8
# Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5)
@ -172,26 +172,55 @@ Source28: 95-kvm-memlock.conf
Source30: kvm-s390x.conf
Source31: kvm-x86.conf
Source36: README.tests
Patch1: 0001-Initial-redhat-build.patch
Patch2: 0002-Enable-disable-devices-for-RHEL.patch
Patch3: 0003-Machine-type-related-general-changes.patch
Patch4: 0004-Add-aarch64-machine-types.patch
Patch5: 0005-Add-ppc64-machine-types.patch
Patch6: 0006-Add-s390x-machine-types.patch
Patch7: 0007-Add-x86_64-machine-types.patch
Patch8: 0008-Enable-make-check.patch
Patch9: 0009-vfio-cap-number-of-devices-that-can-be-assigned.patch
Patch10: 0010-Add-support-statement-to-help-output.patch
Patch11: 0011-Use-qemu-kvm-in-documentation-instead-of-qemu-system.patch
Patch12: 0012-qcow2-Deprecation-warning-when-opening-v2-images-rw.patch
Patch13: kvm-vdpa-use-first-queue-SVQ-state-for-CVQ-default.patch
Patch14: kvm-vdpa-export-vhost_vdpa_set_vring_ready.patch
Patch15: kvm-vdpa-rename-vhost_vdpa_net_load-to-vhost_vdpa_net_cv.patch
Patch16: 0016-vdpa-move-vhost_vdpa_set_vring_ready-to-the-caller.patch
Patch17: kvm-vdpa-remove-net-cvq-migration-blocker.patch
Patch18: 0018-Add-machine-types-compat-bits.patch
Patch19: 0019-Fix-x86-machine-type-compatibility-for-qemu-kvm-8.1..patch
Patch0003: kvm-fix-clang17-s390x.patch
Patch0004: 0004-Initial-redhat-build.patch
Patch0005: 0005-Enable-disable-devices-for-RHEL.patch
Patch0006: 0006-Machine-type-related-general-changes.patch
Patch0007: 0007-Add-aarch64-machine-types.patch
Patch0008: 0008-Add-ppc64-machine-types.patch
Patch0009: 0009-Add-s390x-machine-types.patch
Patch0010: 0010-Add-x86_64-machine-types.patch
Patch0011: 0011-Enable-make-check.patch
Patch0012: 0012-vfio-cap-number-of-devices-that-can-be-assigned.patch
Patch0013: 0013-Add-support-statement-to-help-output.patch
Patch0014: 0014-Use-qemu-kvm-in-documentation-instead-of-qemu-system.patch
Patch0015: 0015-qcow2-Deprecation-warning-when-opening-v2-images-rw.patch
Patch0016: 0016-vdpa-use-first-queue-SVQ-state-for-CVQ-default.patch
Patch0017: 0017-vdpa-export-vhost_vdpa_set_vring_ready.patch
Patch0018: 0018-vdpa-rename-vhost_vdpa_net_load-to-vhost_vdpa_net_cv.patch
Patch0019: 0019-vdpa-move-vhost_vdpa_set_vring_ready-to-the-caller.patch
Patch0020: 0020-vdpa-remove-net-cvq-migration-blocker.patch
Patch0021: 0021-Add-machine-types-compat-bits.patch
Patch0022: 0022-Fix-x86-machine-type-compatibility-for-qemu-kvm-8.1..patch
# For RHEL-832 - qemu-kvm crashed when migrating guest with failover vf
Patch23: kvm-virtio-Drop-out-of-coroutine-context-in-virtio_load.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch24: kvm-migration-Fix-race-that-dest-preempt-thread-close-to.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch25: kvm-migration-Fix-possible-race-when-setting-rp_state.er.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch26: kvm-migration-Fix-possible-races-when-shutting-down-the-.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch27: kvm-migration-Fix-possible-race-when-shutting-down-to_ds.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch28: kvm-migration-Remove-redundant-cleanup-of-postcopy_qemuf.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch29: kvm-migration-Consolidate-return-path-closing-code.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch30: kvm-migration-Replace-the-return-path-retry-logic.patch
# For RHEL-11219 - migration tests failing for RHEL 9.4 sometimes
Patch31: kvm-migration-Move-return-path-cleanup-to-main-migration.patch
# For RHEL-7360 - Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk
Patch32: kvm-file-posix-Clear-bs-bl.zoned-on-error.patch
# For RHEL-7360 - Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk
Patch33: kvm-file-posix-Check-bs-bl.zoned-for-zone-info.patch
# For RHEL-7360 - Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk
Patch34: kvm-file-posix-Fix-zone-update-in-I-O-error-path.patch
# For RHEL-7360 - Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk
Patch35: kvm-file-posix-Simplify-raw_co_prw-s-out-zone-code.patch
# For RHEL-7360 - Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk
Patch36: kvm-tests-file-io-error-New-test.patch
%if %{have_clang}
BuildRequires: clang
@ -1251,6 +1280,35 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%endif
%changelog
* Mon Oct 16 2023 Miroslav Rezanina <mrezanin@redhat.com> - 8.1.0-3
- kvm-migration-Fix-race-that-dest-preempt-thread-close-to.patch [RHEL-11219]
- kvm-migration-Fix-possible-race-when-setting-rp_state.er.patch [RHEL-11219]
- kvm-migration-Fix-possible-races-when-shutting-down-the-.patch [RHEL-11219]
- kvm-migration-Fix-possible-race-when-shutting-down-to_ds.patch [RHEL-11219]
- kvm-migration-Remove-redundant-cleanup-of-postcopy_qemuf.patch [RHEL-11219]
- kvm-migration-Consolidate-return-path-closing-code.patch [RHEL-11219]
- kvm-migration-Replace-the-return-path-retry-logic.patch [RHEL-11219]
- kvm-migration-Move-return-path-cleanup-to-main-migration.patch [RHEL-11219]
- kvm-file-posix-Clear-bs-bl.zoned-on-error.patch [RHEL-7360]
- kvm-file-posix-Check-bs-bl.zoned-for-zone-info.patch [RHEL-7360]
- kvm-file-posix-Fix-zone-update-in-I-O-error-path.patch [RHEL-7360]
- kvm-file-posix-Simplify-raw_co_prw-s-out-zone-code.patch [RHEL-7360]
- kvm-tests-file-io-error-New-test.patch [RHEL-7360]
- Resolves: RHEL-11219
(migration tests failing for RHEL 9.4 sometimes)
- Resolves: RHEL-7360
(Qemu Core Dumped When Writing Larger Size Than The Size of A Data Disk)
* Mon Oct 02 2023 Miroslav Rezanina <mrezanin@redhat.com> - 8.1.0-2
- kvm-virtio-Drop-out-of-coroutine-context-in-virtio_load.patch [RHEL-832]
- Resolves: RHEL-832
(qemu-kvm crashed when migrating guest with failover vf)
* Mon Sep 04 2023 Miroslav Rezanina <mrezanin@redhat.com> - 8.1.0-1
- Rebase to QEMU 8.1 [RHEL-870]
- Resolves: RHEL-870
(Rebase qemu-kvm to QEMU 8.1.0)
* Thu Aug 24 2023 Miroslav Rezanina <mrezanin@redhat.com> - 8.0.0-13
- kvm-vdpa-return-errno-in-vhost_vdpa_get_vring_group-erro.patch [RHEL-923]
- kvm-vdpa-move-CVQ-isolation-check-to-net_init_vhost_vdpa.patch [RHEL-923]