ebasov
/
qemu-kvm
forked from rpms/qemu-kvm
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

import qemu-kvm-6.2.0-32.module+el8.8.0+18361+9f407f6e

Browse Source
i8-stream-rhel imports/c8-stream-rhel/qemu-kvm-6.2.0-32.module+el8.8.0+18361+9f407f6e
CentOS Sources 2 years ago committed by MSVSphere Packaging Team
commit 7b96ed6422
284 changed files with 59108 additions and 0 deletions
Show Stats Download Patch File Download Diff File

1
.gitignore vendored
Unescape View File

@ -0,0 +1 @@
SOURCES/qemu-6.2.0.tar.xz

1
.qemu-kvm.metadata
Unescape View File

@ -0,0 +1 @@
68cd61a466170115b88817e2d52db2cd7a92f43a SOURCES/qemu-6.2.0.tar.xz

17931
SOURCES/0001-redhat-Adding-slirp-to-the-exploded-tree.patch
View File

File diff suppressed because it is too large Load Diff

351
SOURCES/0005-Initial-redhat-build.patch
Unescape View File

@ -0,0 +1,351 @@
From 19ce5ff93ddd6b8a998348f2a5f59f603c5e11b7 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 12 Oct 2018 07:31:11 +0200
Subject: Initial redhat build
This patch introduces redhat build structure in redhat subdirectory. In addition,
several issues are fixed in QEMU tree:
- Change of app name for sasl_server_init in VNC code from qemu to qemu-kvm
- As we use qemu-kvm as name in all places, this is updated to be consistent
- Man page renamed from qemu to qemu-kvm
- man page is installed using make install so we have to fix it in qemu tree
This rebase includes changes up to qemu-kvm-6.1.0-5.el9
Rebase notes (3.1.0):
- added new configure options
Rebase notes (4.0.0):
- Added dependency to perl-Test-Harness (upstream)
- Added dependency to python3-sphinx (upstream)
- Change location of icons (upstream)
- Remove .desktop file (added upstream)
- Added qemu-trace-stap (added upstream)
- Removed elf2dmp (added upstream)
- Remove .buildinfo
- Added pvh.bin rom (added upstream)
- Added interop documentation files
- Use python module instead of qemu.py (upstream)
Rebase notes (4.1.0):
- Remove edk2 files generated by build
- Switch to rhel-8.1-candidate build target
- Remove specs documentation
- Switched from libssh2 to libssh
- Add rc0 tarball usage hacks
- Added BuildRequires for wget, rpm-build and python3-sphinx
- Removed new unpacked files
- Update configure line to use new options
Rebase notes (4.2.0):
- Disable iotest run during make check
- README renamed to README.rst (upstream)
- Removed ui-spice-app.so
- Added relevant changes from "505f7f4 redhat: Adding slirp to the exploded tree"
- Removed qemu-ga.8 install from spec file - installed by make
- Removed spapr-rtas.bin (upstream)
- Require newer SLOF (20191022)
Rebase notes (5.1.0):
- Use python3 for virtio_seg_max_adjust.py test
- Removed qemu-trace-stap shebang from spec file
- Added virtiofsd.1 (upstream)
- Use out-of-tree build
- New documentation structure (upstream)
- Update local build
- Removing installed qemu-storage-daemon (added upstream)
- Removing opensbi-riscv32-sifive_u-fw_jump.bin (added upstream)
- Disable iotests (moved from Enable make check commit)
- Added missing configure options
- Reorder configure options
- qemu-pr-helper moved to /usr/libexec/ (upstream)
- Added submodules for usb-redir, smartcard-reader and qxl display (upstream)
- Added setting rc version in Makefile for build
- removed --disable-vxhs configure option (removed upstream)
- bumped required libusbx-devel version to 1.0.23
- bumped libfdt version to 1.6.0
Rebase notes (5.2.0 rc0):
- Move libfdt dependency to qemu-kvm-core
- Move manpage rename from Makefile to spec file
- rename with-confsuffix configure option to with-suffix (upstream)
- Bump libusbx Requires version to 1.0.234
- Manual copy of keymaps in spec file (BZ 1875217)
- Removed /usr/share/qemu-kvm/npcm7xx_bootrom.bin, considering it
unpackaged for now.
- Removed /usr/share/qemu-kvm/qboot.rom, considering unpackaged.
- Added build dependency for meson and ninja-build
- hw/s390/s390-pci-vfio.c hack - set NULL for g_autofree variables
- Removed Chanelog (upstream)
- Fix in directory used for docs (upstream add %name so we do not pass it in configure)
- Package various .so as part of qemu-kvm-core package.
Rebase notes (5.2.0 rc2):
- Added fix for dtrace build on RHEL 8.4.0
Rebase notes (5.2.0 rc3):
- Added man page for qemu-pr-helper
- Added new configure options
- Update qemu-kiwi patches to v4
Rebase notes (6.0.0):
- update tracetool usage in spec file
- remove qemu-storage-daemon-qmp-ref man page
- remove qemu-storage-daemon man page
- Added devel documentation
- do not package virtfs-proxy-helper files
- Use --with-git-submodules instead of --(enable|disable)-git-update
- Minor build fixes for sending upstream
- g_autofree initialization fixed upstream
- Updated rc information usage
- do not package package hw-s390x-virtio-gpu-ccw.so
- Disable new switch options
Rebase notes (6.1.0):
- Fix warning issue in block.c
- Download tarball from dist-git cache
- Removed sheepdog driver
- Added new display modules:
- hw-display-virtio-gpu-gl.so
- hw-display-virtio-gpu-pci-gl.so
- hw-display-virtio-vga-gl.so
- sasl fix moved from ui/vnc.c to ui/vnc-auth-sasl.c
- Added accel-qtest-%{kvm_target} and accel-tcg-%{kvm_target}
- Added about docs
- Use -q option for setup
- Added hw-usb-host.so
- Disable new options (bpf, nvmm, slirp-smbd)
Rebase notes (6.2.0):
- Using internal meson
- removed --disable-jemalloc and --disable-tcmalloc configure options
- added audio-oss.so
- added fdt requirement for x86_64
- tests/acceptance renamed to tests/avocado
- added multiboot_dma.bin
- Removed conflict relics
- Updated configure options
Merged patches (3.1.0):
- 01f0c9f RHEL8: Add disable configure options to qemu spec file
- Spec file cleanups
Merged patches (4.0.0):
- aa4297c Add edk2 Requires to qemu-kvm
- d124ff5779 Fixing brew build target
- eb204b5 Introduce the qemu-kvm-tests rpm
- 223cf0c Load kvm module during boot (partial)
Merged patches (4.1.0):
- ebb6e97 redhat: Fix LOCALVERSION creation
- b0ab0cc redhat: enable tpmdev passthrough (not disabling tests)
- 7cb3c4a Enable libpmem to support nvdimm
- 8943607 qemu-kvm.spec: bump libseccomp >= 2.4.0
- 27b7c44 rh: set CONFIG_BOCHS_DISPLAY=y for x86 (partial)
- e1fe9fe x86_64-rh-devices: enable TPM emulation (partial)
Merged patches (4.2.0):
- 69e1fb2 enable virgla
- d4f6115 enable virgl, for real this time ...
Merged patches (5.1.0):
- 5edf6bd Add support for rh-brew-module
- f77d52d redhat: ship virtiofsd vhost-user device backend
- 63f12d4 redhat: Always use module build target for rh-brew (modified)
- 9b1e140 redhat: updating the modular target
- 44b8bd0 spec: Fix python shenigans for tests
Merged patches (5.2.0 rc0):
- 9238ce7 Add support for simpletrace
- 5797cff Remove explicit glusterfs-api dependency
- fd62478 disable virgl
- 0205018 redhat: link /etc/qemu-ga/fsfreeze-hook to /etc/qemu-kvm/
- 3645097 redhat: Make all generated so files executable (not only block-*)
Merged patches (5.2.0 rc2):
- pjw 99657 redhat: introduces disable_everything macro into the configure call
- pjw 99659 redhat: scripts/extract_build_cmd.py - Avoid listing empty lines
- pjw 99658 redhat: Fixing rh-local build
- pjw 99660 redhat: Add qemu-kiwi subpackage
- d2e59ce redhat: add (un/pre)install systemd hooks for qemu-ga
Merged patches (5.2.0 rc3):
- pjw 99887 - redhat: allow Makefile rh-prep builddep to fail
- pjw 99885 - redhat: adding rh-rpm target
Merged patches (6.0.0):
- 5ab9954a3b spec: find system python via meson
- cd0f7db11f build-system: use b_staticpic=false
- 80d2dec42c udev-kvm-check: remove the "exceeded subscription limit" message
- 38959d51c0 redhat: Allow make to inherit params from parent make for rh-local
- 1e0cfe458f redhat: moving all documentation files to qemu-kvm-docs
- d7a594d02b redhat: makes qemu respect system's crypto profile
- e2bbf1572b spec: Package qemu-storage-daemon
- 92f10993ba spec: ui-spice sub-package
- 8931e46069 spec: ui-opengl sub-package
Merged patches (6.1.0):
- 7bb57541b3 redhat: Install the s390-netboot.img that we've built
- b4a8531f41 redhat: Fix "unversioned Obsoletes" warning
- 141a1693c7 redhat: Move qemu-kvm-docs dependency to qemu-kvm
- d75f59c6f9 redhat: introducting qemu-kvm-hw-usbredir
- a934d8bf44 redhat: use the standard vhost-user JSON path
Merged patches (6.2.0):
- 4f3f04bbb6 spec: Remove qemu-kiwi build
---
.gitignore | 1 +
.gitlab-ci.yml | 24 -
.gitlab/issue_templates/bug.md | 64 -
.gitlab/issue_templates/feature_request.md | 32 -
README.systemtap | 43 +
meson.build | 4 +-
redhat/Makefile | 90 +
redhat/Makefile.common | 48 +
redhat/README.tests | 39 +
redhat/qemu-kvm.spec.template | 3896 ++++++++++++++++++++
redhat/scripts/extract_build_cmd.py | 5 +-
redhat/scripts/process-patches.sh | 20 +-
redhat/scripts/tarball_checksum.sh | 2 +-
redhat/udev-kvm-check.c | 19 +-
scripts/qemu-guest-agent/fsfreeze-hook | 2 +-
scripts/systemtap/conf.d/qemu_kvm.conf | 4 +
scripts/systemtap/script.d/qemu_kvm.stp | 1 +
tests/check-block.sh | 2 +
ui/vnc-auth-sasl.c | 2 +-
19 files changed, 4142 insertions(+), 156 deletions(-)
delete mode 100644 .gitlab-ci.yml
delete mode 100644 .gitlab/issue_templates/bug.md
delete mode 100644 .gitlab/issue_templates/feature_request.md
create mode 100644 README.systemtap
create mode 100644 redhat/Makefile
create mode 100644 redhat/Makefile.common
create mode 100644 redhat/README.tests
create mode 100644 redhat/qemu-kvm.spec.template
create mode 100644 scripts/systemtap/conf.d/qemu_kvm.conf
create mode 100644 scripts/systemtap/script.d/qemu_kvm.stp
diff --git a/README.systemtap b/README.systemtap
new file mode 100644
index 0000000000..ad913fc990
--- /dev/null
+++ b/README.systemtap
@@ -0,0 +1,43 @@
+QEMU tracing using systemtap-initscript
+---------------------------------------
+
+You can capture QEMU trace data all the time using systemtap-initscript. This
+uses SystemTap's flight recorder mode to trace all running guests to a
+fixed-size buffer on the host. Old trace entries are overwritten by new
+entries when the buffer size wraps.
+
+1. Install the systemtap-initscript package:
+ # yum install systemtap-initscript
+
+2. Install the systemtap scripts and the conf file:
+ # cp /usr/share/qemu-kvm/systemtap/script.d/qemu_kvm.stp /etc/systemtap/script.d/
+ # cp /usr/share/qemu-kvm/systemtap/conf.d/qemu_kvm.conf /etc/systemtap/conf.d/
+
+The set of trace events to enable is given in qemu_kvm.stp. This SystemTap
+script can be customized to add or remove trace events provided in
+/usr/share/systemtap/tapset/qemu-kvm-simpletrace.stp.
+
+SystemTap customizations can be made to qemu_kvm.conf to control the flight
+recorder buffer size and whether to store traces in memory only or disk too.
+See stap(1) for option documentation.
+
+3. Start the systemtap service.
+　# service systemtap start qemu_kvm
+
+4. Make the service start at boot time.
+　# chkconfig systemtap on
+
+5. Confirm that the service works.
+ # service systemtap status qemu_kvm
+ qemu_kvm is running...
+
+When you want to inspect the trace buffer, perform the following steps:
+
+1. Dump the trace buffer.
+ # staprun -A qemu_kvm >/tmp/trace.log
+
+2. Start the systemtap service because the preceding step stops the service.
+ # service systemtap start qemu_kvm
+
+3. Translate the trace record to readable format.
+ # /usr/share/qemu-kvm/simpletrace.py --no-header /usr/share/qemu-kvm/trace-events /tmp/trace.log
diff --git a/meson.build b/meson.build
index 96de1a6ef9..5f6ba86dbb 100644
--- a/meson.build
+++ b/meson.build
@@ -2108,7 +2108,9 @@ if capstone_opt == 'internal'
# Include all configuration defines via a header file, which will wind up
# as a dependency on the object file, and thus changes here will result
# in a rebuild.
- '-include', 'capstone-defs.h'
+ '-include', 'capstone-defs.h',
+
+ '-Wp,-D_GLIBCXX_ASSERTIONS',
]
libcapstone = static_library('capstone',
diff --git a/scripts/qemu-guest-agent/fsfreeze-hook b/scripts/qemu-guest-agent/fsfreeze-hook
index 13aafd4845..e9b84ec028 100755
--- a/scripts/qemu-guest-agent/fsfreeze-hook
+++ b/scripts/qemu-guest-agent/fsfreeze-hook
@@ -8,7 +8,7 @@
# request, it is issued with "thaw" argument after filesystem is thawed.
LOGFILE=/var/log/qga-fsfreeze-hook.log
-FSFREEZE_D=$(dirname -- "$0")/fsfreeze-hook.d
+FSFREEZE_D=$(dirname -- "$(realpath $0)")/fsfreeze-hook.d
# Check whether file $1 is a backup or rpm-generated file and should be ignored
is_ignored_file() {
diff --git a/scripts/systemtap/conf.d/qemu_kvm.conf b/scripts/systemtap/conf.d/qemu_kvm.conf
new file mode 100644
index 0000000000..372d8160a4
--- /dev/null
+++ b/scripts/systemtap/conf.d/qemu_kvm.conf
@@ -0,0 +1,4 @@
+# Force load uprobes (see BZ#1118352)
+stap -e 'probe process("/usr/libexec/qemu-kvm").function("main") { printf("") }' -c true
+
+qemu_kvm_OPT="-s4" # per-CPU buffer size, in megabytes
diff --git a/scripts/systemtap/script.d/qemu_kvm.stp b/scripts/systemtap/script.d/qemu_kvm.stp
new file mode 100644
index 0000000000..c04abf9449
--- /dev/null
+++ b/scripts/systemtap/script.d/qemu_kvm.stp
@@ -0,0 +1 @@
+probe qemu.kvm.simpletrace.handle_qmp_command,qemu.kvm.simpletrace.monitor_protocol_*,qemu.kvm.simpletrace.migrate_set_state {}
diff --git a/tests/check-block.sh b/tests/check-block.sh
index f86cb863de..6d38340d49 100755
--- a/tests/check-block.sh
+++ b/tests/check-block.sh
@@ -69,6 +69,8 @@ else
fi
fi
+exit 0
+
cd tests/qemu-iotests
# QEMU_CHECK_BLOCK_AUTO is used to disable some unstable sub-tests
diff --git a/ui/vnc-auth-sasl.c b/ui/vnc-auth-sasl.c
index 47fdae5b21..2a950caa2a 100644
--- a/ui/vnc-auth-sasl.c
+++ b/ui/vnc-auth-sasl.c
@@ -42,7 +42,7 @@
bool vnc_sasl_server_init(Error **errp)
{
- int saslErr = sasl_server_init(NULL, "qemu");
+ int saslErr = sasl_server_init(NULL, "qemu-kvm");
if (saslErr != SASL_OK) {
error_setg(errp, "Failed to initialize SASL auth: %s",
--
2.27.0

795
SOURCES/0006-Enable-disable-devices-for-RHEL.patch
Unescape View File

@ -0,0 +1,795 @@
From 3d5a82d172345d17e300672909835262ff9dc917 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 2 Sep 2020 09:11:07 +0200
Subject: Enable/disable devices for RHEL
This commit adds all changes related to changes in supported devices.
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
Rebase notes (qemu 3.1.0)
- spapr_rng disabled in default_config
- new hyperv.mak in default configs
- Move changes from x86_64-softmmu.mak to i386-softmmu.mak
- Added CONFIG_VIRTIO_MMIO to aarch64-softmmu.mak
- Removed config_vga_isa.c changes as no longer needed
- Removed new devices
Rebase notes (4.0.0):
- Added CONFIG_PCI_EXPRESS_GENERIC_BRIDGE for aarch64-softmmu.mak
- Added CONFIG_ARM_VIRT for aarch64-softmmu.mak
- Switch to KConfig (upstream)
- Using device whitelist + without-defualt-devices option
Rebase notes (4.1.0):
- Added CONFIG_USB_OHCI_PCI for ppc64
- Added CONFIG_XIVE_KVM for ppc64
- Added CONFIG_ACPI_PCI for x86_64
- Added CONFIG_SEMIHOSTING for aarch64
- Cleanup aarch64 devices
- Do not build a15mpcore.c
- Removed ide-isa.c stub file
- Use CONFIG_USB_EHCI_PCI on x86_64 (new upstream)
Rebase notes (4.2.0-rc0):
- Use conditional build for isa-superio.c (upstream change)
- Rename PCI_PIIX to PCI_I440FX (upstream change)
Rebase notes (4.2.0-rc3):
- Disabled ccid-card-emulated (patch 92566)
- Disabled vfio-pci-igd-lpc-bridge (patch 92565)
Rebase notes (5.1.0):
- added CONFIG_PCI_EXPRESS on ppc64 (due to upstream dependency)
- Added CONFIG_NVDIMM
- updated cortex-15 disabling to upstream code
- Add CONFIG_ACPI_APEI for aarch64
- removed obsolete hw/bt/Makefile.objs chunk
- removed unnecessary changes in target/i386/cpu.c
Rebase notes (5.2.0 rc0):
- Added CONFIG_USB_XHCI_PCI on aarch64 ppc64 and x86_64
- remove vl.c hack for no hpet
- Enable CONFIG_PTIMER for aarch64
- Do not package hw-display-virtio-gpu.so on s390x
Rebase notes (5.2.0 rc1):
- Added CONFIG_ARM_GIC for aarch64 (required for build)
Rebase notes (weekly-210113):
- Removed XICS_KVM, XICS_SPAPR, XIVE_KVM and XIVE_SPAPR config (removed upstream)
Rebase notes (weekly-210120):
- Add CONFIG_ARM_COMPATIBLE_SEMIHOSTING option
Rebase notes (weekly-210203):
- Rename CONFIG_PVPANIC to CONFIG_PVPANIC_ISA
Rebase notes (weekly-210317):
- Add new USB_STORAGE_CORE and USB_STORAGE_CLASSIC config for ppc64 and x86_64
- Update disabling TCG cpus for AArch64
Rebase notes (weekly-210519):
- Do not use CONFIG_SPICE and CONFIG_OPENGL in default configs
Rebase notes (weekly-210623):
- Add CONFIG_TPM for archs with used TPM functionality
Rebase notes (weekly-210714):
- default_configs moved to configs
Rebase notes (6.1.0 rc2):
- Use --with-device-ARCH configure option to use redhat config files
Rebase notes (6.2.0 rc3):
- Do not remove -no-hpet documentation
Merged patches (qemu 3.1.0):
- d51e082 Re-enable CONFIG_HYPERV_TESTDEV
- 4b889f3 Declare cirrus-vga as deprecated
- b579d32 Do not build bluetooth support
- 3eef52a Disable CONFIG_IPMI and CONFIG_I2C for ppc64
- 9caf292 Disable CONFIG_CAN_BUS and CONFIG_CAN_SJA1000
Merged patches (4.1.0):
- 20a51f6 fdc: Revert downstream disablement of device "floppy"
- f869cc0 fdc: Restrict floppy controllers to RHEL-7 machine types
- 5909721 aarch64: Compile out IOH3420
- 27b7c44 rh: set CONFIG_BOCHS_DISPLAY=y for x86 (partial)
- 495a27d x86_64-rh-devices: add missing TPM passthrough
- e1fe9fe x86_64-rh-devices: enable TPM emulation (partial)
Merged patches (4.2.0):
- f7587dd RHEL: disable hostmem-memfd
Merged patches (5.1.0):
- 4543a3c i386: Remove cpu64-rhel6 CPU model
- 96533 aarch64: Remove tcg cpu types (pjw commit)
- 559d589 Revert "RHEL: disable hostmem-memfd"
- 441128e enable ramfb
Merged patches (5.2.0 rc0):
- f70eb50 RHEL-only: Enable vTPM for POWER in downstream configs
- 69d8ae7 redhat: fix 5.0 rebase missing ISA TPM TIS
- 8310f89 RHEL-only: Enable vTPM for ARM in downstream configs
- 4a8ccfd Disable TPM passthrough backend on ARM
Merged patches (6.0.0):
- ff817df9e3 config: enable VFIO_CCW
- 70d3924521 redhat: Add some devices for exporting upstream machine types
- without machine type chunks
- efac91b2b4 default-configs: Enable vhost-user-blk
Merged patches (weekly-210630):
- 59a178acff disable CONFIG_USB_STORAGE_BOT
Merged patches (6.1.0 rc2):
- 86f0025f16 aarch64: Add USB storage devices
---
.../aarch64-softmmu/aarch64-rh-devices.mak | 31 ++++++
.../ppc64-softmmu/ppc64-rh-devices.mak | 36 ++++++
configs/devices/rh-virtio.mak | 10 ++
.../s390x-softmmu/s390x-rh-devices.mak | 16 +++
.../x86_64-softmmu/x86_64-rh-devices.mak | 104 ++++++++++++++++++
.../x86_64-upstream-devices.mak | 4 +
hw/acpi/ich9.c | 4 +-
hw/arm/meson.build | 2 +-
hw/block/fdc.c | 10 ++
hw/char/parallel.c | 9 ++
hw/cpu/meson.build | 5 +-
hw/display/cirrus_vga.c | 3 +
hw/ide/piix.c | 5 +-
hw/input/pckbd.c | 2 +
hw/net/e1000.c | 2 +
hw/ppc/spapr_cpu_core.c | 2 +
hw/timer/hpet.c | 8 ++
hw/usb/meson.build | 2 +-
redhat/qemu-kvm.spec.template | 9 +-
target/arm/cpu_tcg.c | 10 ++
target/ppc/cpu-models.c | 10 ++
target/s390x/cpu_models_sysemu.c | 3 +
target/s390x/kvm/kvm.c | 8 ++
23 files changed, 286 insertions(+), 9 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
create mode 100644 configs/devices/s390x-softmmu/s390x-rh-devices.mak
create mode 100644 configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
create mode 100644 configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak
diff --git a/configs/devices/aarch64-softmmu/aarch64-rh-devices.mak b/configs/devices/aarch64-softmmu/aarch64-rh-devices.mak
new file mode 100644
index 0000000000..0d4f9e6e4b
--- /dev/null
+++ b/configs/devices/aarch64-softmmu/aarch64-rh-devices.mak
@@ -0,0 +1,31 @@
+include ../rh-virtio.mak
+
+CONFIG_ARM_GIC_KVM=y
+CONFIG_ARM_GIC=y
+CONFIG_ARM_SMMUV3=y
+CONFIG_ARM_V7M=y
+CONFIG_ARM_VIRT=y
+CONFIG_EDID=y
+CONFIG_PCIE_PORT=y
+CONFIG_PCI_DEVICES=y
+CONFIG_PCI_TESTDEV=y
+CONFIG_PFLASH_CFI01=y
+CONFIG_SCSI=y
+CONFIG_SEMIHOSTING=y
+CONFIG_USB=y
+CONFIG_USB_XHCI=y
+CONFIG_USB_XHCI_PCI=y
+CONFIG_USB_STORAGE_CORE=y
+CONFIG_USB_STORAGE_CLASSIC=y
+CONFIG_VFIO=y
+CONFIG_VFIO_PCI=y
+CONFIG_VIRTIO_MMIO=y
+CONFIG_VIRTIO_PCI=y
+CONFIG_XIO3130=y
+CONFIG_NVDIMM=y
+CONFIG_ACPI_APEI=y
+CONFIG_TPM=y
+CONFIG_TPM_EMULATOR=y
+CONFIG_TPM_TIS_SYSBUS=y
+CONFIG_PTIMER=y
+CONFIG_ARM_COMPATIBLE_SEMIHOSTING=y
diff --git a/configs/devices/ppc64-softmmu/ppc64-rh-devices.mak b/configs/devices/ppc64-softmmu/ppc64-rh-devices.mak
new file mode 100644
index 0000000000..73e3ee0293
--- /dev/null
+++ b/configs/devices/ppc64-softmmu/ppc64-rh-devices.mak
@@ -0,0 +1,36 @@
+include ../rh-virtio.mak
+
+CONFIG_DIMM=y
+CONFIG_MEM_DEVICE=y
+CONFIG_NVDIMM=y
+CONFIG_PCI=y
+CONFIG_PCI_DEVICES=y
+CONFIG_PCI_TESTDEV=y
+CONFIG_PCI_EXPRESS=y
+CONFIG_PSERIES=y
+CONFIG_SCSI=y
+CONFIG_SPAPR_VSCSI=y
+CONFIG_TEST_DEVICES=y
+CONFIG_USB=y
+CONFIG_USB_OHCI=y
+CONFIG_USB_OHCI_PCI=y
+CONFIG_USB_SMARTCARD=y
+CONFIG_USB_STORAGE_CORE=y
+CONFIG_USB_STORAGE_CLASSIC=y
+CONFIG_USB_XHCI=y
+CONFIG_USB_XHCI_NEC=y
+CONFIG_USB_XHCI_PCI=y
+CONFIG_VFIO=y
+CONFIG_VFIO_PCI=y
+CONFIG_VGA=y
+CONFIG_VGA_PCI=y
+CONFIG_VHOST_USER=y
+CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VGA=y
+CONFIG_WDT_IB6300ESB=y
+CONFIG_XICS=y
+CONFIG_XIVE=y
+CONFIG_TPM=y
+CONFIG_TPM_SPAPR=y
+CONFIG_TPM_EMULATOR=y
+CONFIG_TPM_PASSTHROUGH=y
diff --git a/configs/devices/rh-virtio.mak b/configs/devices/rh-virtio.mak
new file mode 100644
index 0000000000..94ede1b5f6
--- /dev/null
+++ b/configs/devices/rh-virtio.mak
@@ -0,0 +1,10 @@
+CONFIG_VIRTIO=y
+CONFIG_VIRTIO_BALLOON=y
+CONFIG_VIRTIO_BLK=y
+CONFIG_VIRTIO_GPU=y
+CONFIG_VIRTIO_INPUT=y
+CONFIG_VIRTIO_INPUT_HOST=y
+CONFIG_VIRTIO_NET=y
+CONFIG_VIRTIO_RNG=y
+CONFIG_VIRTIO_SCSI=y
+CONFIG_VIRTIO_SERIAL=y
diff --git a/configs/devices/s390x-softmmu/s390x-rh-devices.mak b/configs/devices/s390x-softmmu/s390x-rh-devices.mak
new file mode 100644
index 0000000000..165c082e87
--- /dev/null
+++ b/configs/devices/s390x-softmmu/s390x-rh-devices.mak
@@ -0,0 +1,16 @@
+include ../rh-virtio.mak
+
+CONFIG_PCI=y
+CONFIG_S390_CCW_VIRTIO=y
+CONFIG_S390_FLIC=y
+CONFIG_S390_FLIC_KVM=y
+CONFIG_SCLPCONSOLE=y
+CONFIG_SCSI=y
+CONFIG_TERMINAL3270=y
+CONFIG_VFIO=y
+CONFIG_VFIO_AP=y
+CONFIG_VFIO_CCW=y
+CONFIG_VFIO_PCI=y
+CONFIG_VHOST_USER=y
+CONFIG_VIRTIO_CCW=y
+CONFIG_WDT_DIAG288=y
diff --git a/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak b/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
new file mode 100644
index 0000000000..ddf036f042
--- /dev/null
+++ b/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
@@ -0,0 +1,104 @@
+include ../rh-virtio.mak
+include x86_64-upstream-devices.mak
+
+CONFIG_AC97=y
+CONFIG_ACPI=y
+CONFIG_ACPI_PCI=y
+CONFIG_ACPI_CPU_HOTPLUG=y
+CONFIG_ACPI_MEMORY_HOTPLUG=y
+CONFIG_ACPI_NVDIMM=y
+CONFIG_ACPI_SMBUS=y
+CONFIG_ACPI_VMGENID=y
+CONFIG_ACPI_X86=y
+CONFIG_ACPI_X86_ICH=y
+CONFIG_AHCI=y
+CONFIG_APIC=y
+CONFIG_APM=y
+CONFIG_BOCHS_DISPLAY=y
+CONFIG_DIMM=y
+CONFIG_E1000E_PCI_EXPRESS=y
+CONFIG_E1000_PCI=y
+CONFIG_EDU=y
+CONFIG_FDC=y
+CONFIG_FDC_SYSBUS=y
+CONFIG_FW_CFG_DMA=y
+CONFIG_HDA=y
+CONFIG_HYPERV=y
+CONFIG_HYPERV_TESTDEV=y
+CONFIG_I2C=y
+CONFIG_I440FX=y
+CONFIG_I8254=y
+CONFIG_I8257=y
+CONFIG_I8259=y
+CONFIG_I82801B11=y
+CONFIG_IDE_CORE=y
+CONFIG_IDE_PCI=y
+CONFIG_IDE_PIIX=y
+CONFIG_IDE_QDEV=y
+CONFIG_IOAPIC=y
+CONFIG_IOH3420=y
+CONFIG_ISA_BUS=y
+CONFIG_ISA_DEBUG=y
+CONFIG_ISA_TESTDEV=y
+CONFIG_LPC_ICH9=y
+CONFIG_MC146818RTC=y
+CONFIG_MEM_DEVICE=y
+CONFIG_NVDIMM=y
+CONFIG_PAM=y
+CONFIG_PC=y
+CONFIG_PCI=y
+CONFIG_PCIE_PORT=y
+CONFIG_PCI_DEVICES=y
+CONFIG_PCI_EXPRESS=y
+CONFIG_PCI_EXPRESS_Q35=y
+CONFIG_PCI_I440FX=y
+CONFIG_PCI_TESTDEV=y
+CONFIG_PCKBD=y
+CONFIG_PCSPK=y
+CONFIG_PC_ACPI=y
+CONFIG_PC_PCI=y
+CONFIG_PFLASH_CFI01=y
+CONFIG_PVPANIC_ISA=y
+CONFIG_PXB=y
+CONFIG_Q35=y
+CONFIG_QXL=y
+CONFIG_RTL8139_PCI=y
+CONFIG_SCSI=y
+CONFIG_SERIAL=y
+CONFIG_SERIAL_ISA=y
+CONFIG_SERIAL_PCI=y
+CONFIG_SEV=y
+CONFIG_SGA=y
+CONFIG_SMBIOS=y
+CONFIG_SMBUS_EEPROM=y
+CONFIG_TEST_DEVICES=y
+CONFIG_USB=y
+CONFIG_USB_EHCI=y
+CONFIG_USB_EHCI_PCI=y
+CONFIG_USB_SMARTCARD=y
+CONFIG_USB_STORAGE_CORE=y
+CONFIG_USB_STORAGE_CLASSIC=y
+CONFIG_USB_UHCI=y
+CONFIG_USB_XHCI=y
+CONFIG_USB_XHCI_NEC=y
+CONFIG_USB_XHCI_PCI=y
+CONFIG_VFIO=y
+CONFIG_VFIO_PCI=y
+CONFIG_VGA=y
+CONFIG_VGA_CIRRUS=y
+CONFIG_VGA_PCI=y
+CONFIG_VHOST_USER=y
+CONFIG_VHOST_USER_BLK=y
+CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VGA=y
+CONFIG_VMMOUSE=y
+CONFIG_VMPORT=y
+CONFIG_VTD=y
+CONFIG_WDT_IB6300ESB=y
+CONFIG_WDT_IB700=y
+CONFIG_XIO3130=y
+CONFIG_TPM=y
+CONFIG_TPM_CRB=y
+CONFIG_TPM_TIS_ISA=y
+CONFIG_TPM_EMULATOR=y
+CONFIG_TPM_PASSTHROUGH=y
diff --git a/configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak b/configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak
new file mode 100644
index 0000000000..2cd20f54d2
--- /dev/null
+++ b/configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak
@@ -0,0 +1,4 @@
+# We need "isa-parallel"
+CONFIG_PARALLEL=y
+# We need "hpet"
+CONFIG_HPET=y
diff --git a/hw/acpi/ich9.c b/hw/acpi/ich9.c
index ebe08ed831..381ef2ddcf 100644
--- a/hw/acpi/ich9.c
+++ b/hw/acpi/ich9.c
@@ -438,8 +438,8 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
static const uint32_t gpe0_len = ICH9_PMIO_GPE0_LEN;
pm->acpi_memory_hotplug.is_enabled = true;
pm->cpu_hotplug_legacy = true;
- pm->disable_s3 = 0;
- pm->disable_s4 = 0;
+ pm->disable_s3 = 1;
+ pm->disable_s4 = 1;
pm->s4_val = 2;
pm->use_acpi_hotplug_bridge = true;
pm->keep_pci_slot_hpc = true;
diff --git a/hw/arm/meson.build b/hw/arm/meson.build
index 721a8eb8be..87ed4dd914 100644
--- a/hw/arm/meson.build
+++ b/hw/arm/meson.build
@@ -31,7 +31,7 @@ arm_ss.add(when: 'CONFIG_VEXPRESS', if_true: files('vexpress.c'))
arm_ss.add(when: 'CONFIG_ZYNQ', if_true: files('xilinx_zynq.c'))
arm_ss.add(when: 'CONFIG_SABRELITE', if_true: files('sabrelite.c'))
-arm_ss.add(when: 'CONFIG_ARM_V7M', if_true: files('armv7m.c'))
+#arm_ss.add(when: 'CONFIG_ARM_V7M', if_true: files('armv7m.c'))
arm_ss.add(when: 'CONFIG_EXYNOS4', if_true: files('exynos4210.c'))
arm_ss.add(when: 'CONFIG_PXA2XX', if_true: files('pxa2xx.c', 'pxa2xx_gpio.c', 'pxa2xx_pic.c'))
arm_ss.add(when: 'CONFIG_DIGIC', if_true: files('digic.c'))
diff --git a/hw/block/fdc.c b/hw/block/fdc.c
index 21d18ac2e3..97fa6de423 100644
--- a/hw/block/fdc.c
+++ b/hw/block/fdc.c
@@ -48,6 +48,8 @@
#include "qom/object.h"
#include "fdc-internal.h"
+#include "hw/boards.h"
+
/********************************************************/
/* debug Floppy devices */
@@ -2337,6 +2339,14 @@ void fdctrl_realize_common(DeviceState *dev, FDCtrl *fdctrl, Error **errp)
FDrive *drive;
static int command_tables_inited = 0;
+ /* Restricted for Red Hat Enterprise Linux: */
+ MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
+ if (!strstr(mc->name, "-rhel7.")) {
+ error_setg(errp, "Device %s is not supported with machine type %s",
+ object_get_typename(OBJECT(dev)), mc->name);
+ return;
+ }
+
if (fdctrl->fallback == FLOPPY_DRIVE_TYPE_AUTO) {
error_setg(errp, "Cannot choose a fallback FDrive type of 'auto'");
return;
diff --git a/hw/char/parallel.c b/hw/char/parallel.c
index b45e67bfbb..e5f108211b 100644
--- a/hw/char/parallel.c
+++ b/hw/char/parallel.c
@@ -29,6 +29,7 @@
#include "chardev/char-parallel.h"
#include "chardev/char-fe.h"
#include "hw/acpi/aml-build.h"
+#include "hw/boards.h"
#include "hw/irq.h"
#include "hw/isa/isa.h"
#include "hw/qdev-properties.h"
@@ -534,6 +535,14 @@ static void parallel_isa_realizefn(DeviceState *dev, Error **errp)
int base;
uint8_t dummy;
+ /* Restricted for Red Hat Enterprise Linux */
+ MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
+ if (strstr(mc->name, "rhel")) {
+ error_setg(errp, "Device %s is not supported with machine type %s",
+ object_get_typename(OBJECT(dev)), mc->name);
+ return;
+ }
+
if (!qemu_chr_fe_backend_connected(&s->chr)) {
error_setg(errp, "Can't create parallel device, empty char device");
return;
diff --git a/hw/cpu/meson.build b/hw/cpu/meson.build
index 9e52fee9e7..bb71c9f3e7 100644
--- a/hw/cpu/meson.build
+++ b/hw/cpu/meson.build
@@ -1,6 +1,7 @@
-softmmu_ss.add(files('core.c', 'cluster.c'))
+#softmmu_ss.add(files('core.c', 'cluster.c'))
+softmmu_ss.add(files('core.c'))
specific_ss.add(when: 'CONFIG_ARM11MPCORE', if_true: files('arm11mpcore.c'))
specific_ss.add(when: 'CONFIG_REALVIEW', if_true: files('realview_mpcore.c'))
specific_ss.add(when: 'CONFIG_A9MPCORE', if_true: files('a9mpcore.c'))
-specific_ss.add(when: 'CONFIG_A15MPCORE', if_true: files('a15mpcore.c'))
+#specific_ss.add(when: 'CONFIG_A15MPCORE', if_true: files('a15mpcore.c'))
diff --git a/hw/display/cirrus_vga.c b/hw/display/cirrus_vga.c
index fdca6ca659..fa1a7eee51 100644
--- a/hw/display/cirrus_vga.c
+++ b/hw/display/cirrus_vga.c
@@ -2945,6 +2945,9 @@ static void pci_cirrus_vga_realize(PCIDevice *dev, Error **errp)
PCIDeviceClass *pc = PCI_DEVICE_GET_CLASS(dev);
int16_t device_id = pc->device_id;
+ warn_report("'cirrus-vga' is deprecated, "
+ "please use a different VGA card instead");
+
/* follow real hardware, cirrus card emulated has 4 MB video memory.
Also accept 8 MB/16 MB for backward compatibility. */
if (s->vga.vram_size_mb != 4 && s->vga.vram_size_mb != 8 &&
diff --git a/hw/ide/piix.c b/hw/ide/piix.c
index ce89fd0aa3..fbcf802b13 100644
--- a/hw/ide/piix.c
+++ b/hw/ide/piix.c
@@ -232,7 +232,8 @@ static void piix3_ide_class_init(ObjectClass *klass, void *data)
k->device_id = PCI_DEVICE_ID_INTEL_82371SB_1;
k->class_id = PCI_CLASS_STORAGE_IDE;
set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
- dc->hotpluggable = false;
+ /* Disabled for Red Hat Enterprise Linux: */
+ dc->user_creatable = false;
}
static const TypeInfo piix3_ide_info = {
@@ -261,6 +262,8 @@ static void piix4_ide_class_init(ObjectClass *klass, void *data)
k->class_id = PCI_CLASS_STORAGE_IDE;
set_bit(DEVICE_CATEGORY_STORAGE, dc->categories);
dc->hotpluggable = false;
+ /* Disabled for Red Hat Enterprise Linux: */
+ dc->user_creatable = false;
}
static const TypeInfo piix4_ide_info = {
diff --git a/hw/input/pckbd.c b/hw/input/pckbd.c
index baba62f357..bc360347ea 100644
--- a/hw/input/pckbd.c
+++ b/hw/input/pckbd.c
@@ -796,6 +796,8 @@ static void i8042_class_initfn(ObjectClass *klass, void *data)
dc->vmsd = &vmstate_kbd_isa;
isa->build_aml = i8042_build_aml;
set_bit(DEVICE_CATEGORY_INPUT, dc->categories);
+ /* Disabled for Red Hat Enterprise Linux: */
+ dc->user_creatable = false;
}
static const TypeInfo i8042_info = {
diff --git a/hw/net/e1000.c b/hw/net/e1000.c
index f5bc81296d..282d01e374 100644
--- a/hw/net/e1000.c
+++ b/hw/net/e1000.c
@@ -1821,6 +1821,7 @@ static const E1000Info e1000_devices[] = {
.revision = 0x03,
.phy_id2 = E1000_PHY_ID2_8254xx_DEFAULT,
},
+#if 0 /* Disabled for Red Hat Enterprise Linux 7 */
{
.name = "e1000-82544gc",
.device_id = E1000_DEV_ID_82544GC_COPPER,
@@ -1833,6 +1834,7 @@ static const E1000Info e1000_devices[] = {
.revision = 0x03,
.phy_id2 = E1000_PHY_ID2_8254xx_DEFAULT,
},
+#endif
};
static void e1000_register_types(void)
diff --git a/hw/ppc/spapr_cpu_core.c b/hw/ppc/spapr_cpu_core.c
index 58e7341cb7..8ba34f6a1d 100644
--- a/hw/ppc/spapr_cpu_core.c
+++ b/hw/ppc/spapr_cpu_core.c
@@ -370,10 +370,12 @@ static const TypeInfo spapr_cpu_core_type_infos[] = {
.instance_size = sizeof(SpaprCpuCore),
.class_size = sizeof(SpaprCpuCoreClass),
},
+#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_SPAPR_CPU_CORE_TYPE("970_v2.2"),
DEFINE_SPAPR_CPU_CORE_TYPE("970mp_v1.0"),
DEFINE_SPAPR_CPU_CORE_TYPE("970mp_v1.1"),
DEFINE_SPAPR_CPU_CORE_TYPE("power5+_v2.1"),
+#endif
DEFINE_SPAPR_CPU_CORE_TYPE("power7_v2.3"),
DEFINE_SPAPR_CPU_CORE_TYPE("power7+_v2.1"),
DEFINE_SPAPR_CPU_CORE_TYPE("power8_v2.0"),
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index 9520471be2..202e032524 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -733,6 +733,14 @@ static void hpet_realize(DeviceState *dev, Error **errp)
int i;
HPETTimer *timer;
+ /* Restricted for Red Hat Enterprise Linux */
+ MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
+ if (strstr(mc->name, "rhel")) {
+ error_setg(errp, "Device %s is not supported with machine type %s",
+ object_get_typename(OBJECT(dev)), mc->name);
+ return;
+ }
+
if (!s->intcap) {
warn_report("Hpet's intcap not initialized");
}
diff --git a/hw/usb/meson.build b/hw/usb/meson.build
index de853d780d..0776ae6a20 100644
--- a/hw/usb/meson.build
+++ b/hw/usb/meson.build
@@ -52,7 +52,7 @@ softmmu_ss.add(when: 'CONFIG_USB_SMARTCARD', if_true: files('dev-smartcard-reade
if cacard.found()
usbsmartcard_ss = ss.source_set()
usbsmartcard_ss.add(when: 'CONFIG_USB_SMARTCARD',
- if_true: [cacard, files('ccid-card-emulated.c', 'ccid-card-passthru.c')])
+ if_true: [cacard, files('ccid-card-passthru.c')])
hw_usb_modules += {'smartcard': usbsmartcard_ss}
endif
diff --git a/target/arm/cpu_tcg.c b/target/arm/cpu_tcg.c
index 13d0e9b195..3826fa5122 100644
--- a/target/arm/cpu_tcg.c
+++ b/target/arm/cpu_tcg.c
@@ -22,6 +22,7 @@
/* CPU models. These are not needed for the AArch64 linux-user build. */
#if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64)
+#if 0 /* Disabled for Red Hat Enterprise Linux */
#if !defined(CONFIG_USER_ONLY) && defined(CONFIG_TCG)
static bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
{
@@ -375,6 +376,7 @@ static void cortex_a9_initfn(Object *obj)
cpu->ccsidr[1] = 0x200fe019; /* 16k L1 icache. */
define_arm_cp_regs(cpu, cortexa9_cp_reginfo);
}
+#endif /* disabled for RHEL */
#ifndef CONFIG_USER_ONLY
static uint64_t a15_l2ctlr_read(CPUARMState *env, const ARMCPRegInfo *ri)
@@ -400,6 +402,7 @@ static const ARMCPRegInfo cortexa15_cp_reginfo[] = {
REGINFO_SENTINEL
};
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void cortex_a7_initfn(Object *obj)
{
ARMCPU *cpu = ARM_CPU(obj);
@@ -445,6 +448,7 @@ static void cortex_a7_initfn(Object *obj)
cpu->ccsidr[2] = 0x711fe07a; /* 4096K L2 unified cache */
define_arm_cp_regs(cpu, cortexa15_cp_reginfo); /* Same as A15 */
}
+#endif /* disabled for RHEL */
static void cortex_a15_initfn(Object *obj)
{
@@ -488,6 +492,7 @@ static void cortex_a15_initfn(Object *obj)
define_arm_cp_regs(cpu, cortexa15_cp_reginfo);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void cortex_m0_initfn(Object *obj)
{
ARMCPU *cpu = ARM_CPU(obj);
@@ -928,6 +933,7 @@ static void arm_v7m_class_init(ObjectClass *oc, void *data)
cc->gdb_core_xml_file = "arm-m-profile.xml";
}
+#endif /* disabled for RHEL */
#ifndef TARGET_AARCH64
/*
@@ -1007,6 +1013,7 @@ static void arm_max_initfn(Object *obj)
#endif /* !TARGET_AARCH64 */
static const ARMCPUInfo arm_tcg_cpus[] = {
+#if 0 /* Disabled for Red Hat Enterprise Linux */
{ .name = "arm926", .initfn = arm926_initfn },
{ .name = "arm946", .initfn = arm946_initfn },
{ .name = "arm1026", .initfn = arm1026_initfn },
@@ -1022,7 +1029,9 @@ static const ARMCPUInfo arm_tcg_cpus[] = {
{ .name = "cortex-a7", .initfn = cortex_a7_initfn },
{ .name = "cortex-a8", .initfn = cortex_a8_initfn },
{ .name = "cortex-a9", .initfn = cortex_a9_initfn },
+#endif /* disabled for RHEL */
{ .name = "cortex-a15", .initfn = cortex_a15_initfn },
+#if 0 /* Disabled for Red Hat Enterprise Linux */
{ .name = "cortex-m0", .initfn = cortex_m0_initfn,
.class_init = arm_v7m_class_init },
{ .name = "cortex-m3", .initfn = cortex_m3_initfn,
@@ -1053,6 +1062,7 @@ static const ARMCPUInfo arm_tcg_cpus[] = {
{ .name = "pxa270-b1", .initfn = pxa270b1_initfn },
{ .name = "pxa270-c0", .initfn = pxa270c0_initfn },
{ .name = "pxa270-c5", .initfn = pxa270c5_initfn },
+#endif /* disabled for RHEL */
#ifndef TARGET_AARCH64
{ .name = "max", .initfn = arm_max_initfn },
#endif
diff --git a/target/ppc/cpu-models.c b/target/ppc/cpu-models.c
index 4baa111713..d779c4d1d5 100644
--- a/target/ppc/cpu-models.c
+++ b/target/ppc/cpu-models.c
@@ -66,6 +66,7 @@
#define POWERPC_DEF(_name, _pvr, _type, _desc) \
POWERPC_DEF_SVR(_name, _desc, _pvr, POWERPC_SVR_NONE, _type)
+#if 0 /* Embedded and 32-bit CPUs disabled for Red Hat Enterprise Linux */
/* Embedded PowerPC */
/* PowerPC 401 family */
POWERPC_DEF("401", CPU_POWERPC_401, 401,
@@ -740,8 +741,10 @@
"PowerPC 7447A v1.2 (G4)")
POWERPC_DEF("7457a_v1.2", CPU_POWERPC_74x7A_v12, 7455,
"PowerPC 7457A v1.2 (G4)")
+#endif
/* 64 bits PowerPC */
#if defined(TARGET_PPC64)
+#if 0 /* Disabled for Red Hat Enterprise Linux */
POWERPC_DEF("970_v2.2", CPU_POWERPC_970_v22, 970,
"PowerPC 970 v2.2")
POWERPC_DEF("970fx_v1.0", CPU_POWERPC_970FX_v10, 970,
@@ -760,6 +763,7 @@
"PowerPC 970MP v1.1")
POWERPC_DEF("power5+_v2.1", CPU_POWERPC_POWER5P_v21, POWER5P,
"POWER5+ v2.1")
+#endif
POWERPC_DEF("power7_v2.3", CPU_POWERPC_POWER7_v23, POWER7,
"POWER7 v2.3")
POWERPC_DEF("power7+_v2.1", CPU_POWERPC_POWER7P_v21, POWER7,
@@ -784,6 +788,7 @@
/* PowerPC CPU aliases */
PowerPCCPUAlias ppc_cpu_aliases[] = {
+#if 0 /* Embedded and 32-bit CPUs disabled for Red Hat Enterprise Linux */
{ "403", "403gc" },
{ "405", "405d4" },
{ "405cr", "405crc" },
@@ -942,12 +947,15 @@ PowerPCCPUAlias ppc_cpu_aliases[] = {
{ "7447a", "7447a_v1.2" },
{ "7457a", "7457a_v1.2" },
{ "apollo7pm", "7457a_v1.0" },
+#endif
#if defined(TARGET_PPC64)
+#if 0 /* Disabled for Red Hat Enterprise Linux */
{ "970", "970_v2.2" },
{ "970fx", "970fx_v3.1" },
{ "970mp", "970mp_v1.1" },
{ "power5+", "power5+_v2.1" },
{ "power5gs", "power5+_v2.1" },
+#endif
{ "power7", "power7_v2.3" },
{ "power7+", "power7+_v2.1" },
{ "power8e", "power8e_v2.1" },
@@ -957,6 +965,7 @@ PowerPCCPUAlias ppc_cpu_aliases[] = {
{ "power10", "power10_v2.0" },
#endif
+#if 0 /* Disabled for Red Hat Enterprise Linux */
/* Generic PowerPCs */
#if defined(TARGET_PPC64)
{ "ppc64", "970fx_v3.1" },
@@ -964,5 +973,6 @@ PowerPCCPUAlias ppc_cpu_aliases[] = {
{ "ppc32", "604" },
{ "ppc", "604" },
{ "default", "604" },
+#endif
{ NULL, NULL }
};
diff --git a/target/s390x/cpu_models_sysemu.c b/target/s390x/cpu_models_sysemu.c
index 05c3ccaaff..6a04ccab1b 100644
--- a/target/s390x/cpu_models_sysemu.c
+++ b/target/s390x/cpu_models_sysemu.c
@@ -36,6 +36,9 @@ static void check_unavailable_features(const S390CPUModel *max_model,
(max_model->def->gen == model->def->gen &&
max_model->def->ec_ga < model->def->ec_ga)) {
list_add_feat("type", unavailable);
+ } else if (model->def->gen < 11 && kvm_enabled()) {
+ /* Older CPU models are not supported on Red Hat Enterprise Linux */
+ list_add_feat("type", unavailable);
}
/* detect missing features if any to properly report them */
diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
index 5b1fdb55c4..c52434985b 100644
--- a/target/s390x/kvm/kvm.c
+++ b/target/s390x/kvm/kvm.c
@@ -2508,6 +2508,14 @@ void kvm_s390_apply_cpu_model(const S390CPUModel *model, Error **errp)
error_setg(errp, "KVM doesn't support CPU models");
return;
}
+
+ /* Older CPU models are not supported on Red Hat Enterprise Linux */
+ if (model->def->gen < 11) {
+ error_setg(errp, "KVM: Unsupported CPU type specified: %s",
+ MACHINE(qdev_get_machine())->cpu_type);
+ return;
+ }
+
prop.cpuid = s390_cpuid_from_cpu_model(model);
prop.ibc = s390_ibc_from_cpu_model(model);
/* configure cpu features indicated via STFL(e) */
--
2.27.0

1071
SOURCES/0007-Machine-type-related-general-changes.patch
View File

File diff suppressed because it is too large Load Diff

405
SOURCES/0008-Add-aarch64-machine-types.patch
Unescape View File

@ -0,0 +1,405 @@
From 670e90f5cbd92189155e079b8c6e2aafdf82d162 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 12:53:31 +0200
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 (4.0.0):
- Use upstream compat handling
Rebase notes (4.1.0-rc0):
- Removed a15memmap (upstream)
- Use virt_flash_create in rhel800_virt_instance_init
Rebase notes (4.2.0-rc0):
- Set numa_mem_supported
Rebase notes (4.2.0-rc3):
- aarch64: Add virt-rhel8.2.0 machine type for ARM (patch 92246)
- aarch64: virt: Allow more than 1TB of RAM (patch 92249)
- aarch64: virt: Allow PCDIMM instantiation (patch 92247)
- aarch64: virt: Enhance the comment related to gic-version (patch 92248)
Rebase notes (5.0.0):
- Set default_ram_id in rhel_machine_class_init
- Added setting acpi properties
Rebase notes (5.1.0):
- Added ras property
- Added to virt_machine_device_unplug_cb to machine type (upstream)
- added mte property (upstream)
Rebase notes (weekly-210210):
- Added support for oem fields to machine type
Rebase notes (weekly-210303):
- Use rhel-8.4.0 hw compat
Rebase notes (6.0.0-rc2):
- renamed oem-id and oem-table-id to x-oem-id and x-oem-table-id
Rebase notes (210623):
- Protect TPM functions by CONFIG_TPM ifdef
Rebase notes (6.1.0-rc0):
- Add support for default_bus_bypass_iommu
Merged patches (4.0.0):
- 7bfdb4c aarch64: Add virt-rhel8.0.0 machine type for ARM
- 3433e69 aarch64: Set virt-rhel8.0.0 max_cpus to 512
- 4d20863 aarch64: Use 256MB ECAM region by default
Merged patches (4.1.0):
- c3e39ef aarch64: Add virt-rhel8.1.0 machine type for ARM
- 59a46d1 aarch64: Allow ARM VIRT iommu option in RHEL8.1 machine
Merged patches (5.2.0 rc0):
- 12990ad hw/arm: Changes to rhel820 machine
- 46d5a79 hw/arm: Introduce rhel_virt_instance_init() helper
- 098954a hw/arm: Add rhel830 machine type
- ee8e99d arm: Set correct max_cpus value on virt-rhel* machine types
- e5edd38 RHEL-only: arm/virt: Allow the TPM_TIS_SYSBUS device dynamic allocation in machvirt
- 6d7ba66 machine types/numa: set numa_mem_supported on old machine types (partialy)
- 25c5644 machine_types/numa: compatibility for auto_enable_numa_with_memdev (partialy)
Merged patches (6.0):
- 078fadb5da AArch64 machine types cleanup
- ea7b7425fa hw/arm/virt: Add 8.4 Machine type
Merged patches (weekly-210609):
- 73b1578882 hw/arm/virt: Add 8.5 machine type
- 5333038d11 hw/arm/virt: Disable PL011 clock migration through hw_compat_rhel_8_3
- 63adb8ae86 arm/virt: Register highmem and gic-version as class properties
Merged patches (weekly-211027):
- 86e3057c0a hw: arm: virt: Add hw_compat_rhel_8_5 to 8.5 machine type
---
hw/arm/virt.c | 226 +++++++++++++++++++++++++++++++++++++++++-
hw/core/machine.c | 2 +
include/hw/arm/virt.h | 8 ++
3 files changed, 235 insertions(+), 1 deletion(-)
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 5de4d9d73b..c77d26ab13 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -79,6 +79,7 @@
#include "hw/char/pl011.h"
#include "qemu/guest-random.h"
+#if 0 /* Disabled for Red Hat Enterprise Linux */
#define DEFINE_VIRT_MACHINE_LATEST(major, minor, latest) \
static void virt_##major##_##minor##_class_init(ObjectClass *oc, \
void *data) \
@@ -105,7 +106,48 @@
DEFINE_VIRT_MACHINE_LATEST(major, minor, true)
#define DEFINE_VIRT_MACHINE(major, minor) \
DEFINE_VIRT_MACHINE_LATEST(major, minor, false)
-
+#endif /* disabled for RHEL */
+
+#define DEFINE_RHEL_MACHINE_LATEST(m, n, s, latest) \
+ static void rhel##m##n##s##_virt_class_init(ObjectClass *oc, \
+ void *data) \
+ { \
+ MachineClass *mc = MACHINE_CLASS(oc); \
+ rhel##m##n##s##_virt_options(mc); \
+ mc->desc = "RHEL " # m "." # n "." # s " ARM Virtual Machine"; \
+ if (latest) { \
+ mc->alias = "virt"; \
+ mc->is_default = 1; \
+ } \
+ } \
+ static const TypeInfo rhel##m##n##s##_machvirt_info = { \
+ .name = MACHINE_TYPE_NAME("virt-rhel" # m "." # n "." # s), \
+ .parent = TYPE_RHEL_MACHINE, \
+ .class_init = rhel##m##n##s##_virt_class_init, \
+ }; \
+ static void rhel##m##n##s##_machvirt_init(void) \
+ { \
+ type_register_static(&rhel##m##n##s##_machvirt_info); \
+ } \
+ type_init(rhel##m##n##s##_machvirt_init);
+
+#define DEFINE_RHEL_MACHINE_AS_LATEST(major, minor, subminor) \
+ DEFINE_RHEL_MACHINE_LATEST(major, minor, subminor, true)
+#define DEFINE_RHEL_MACHINE(major, minor, subminor) \
+ DEFINE_RHEL_MACHINE_LATEST(major, minor, subminor, false)
+
+/* This variable is for changes to properties that are RHEL specific,
+ * different to the current upstream and to be applied to the latest
+ * machine type.
+ */
+GlobalProperty arm_rhel_compat[] = {
+ {
+ .driver = "virtio-net-pci",
+ .property = "romfile",
+ .value = "",
+ },
+};
+const size_t arm_rhel_compat_len = G_N_ELEMENTS(arm_rhel_compat);
/* Number of external interrupt lines to configure the GIC with */
#define NUM_IRQS 256
@@ -2180,6 +2222,7 @@ static void machvirt_init(MachineState *machine)
qemu_add_machine_init_done_notifier(&vms->machine_done);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static bool virt_get_secure(Object *obj, Error **errp)
{
VirtMachineState *vms = VIRT_MACHINE(obj);
@@ -2207,6 +2250,7 @@ static void virt_set_virt(Object *obj, bool value, Error **errp)
vms->virt = value;
}
+#endif /* disabled for RHEL */
static bool virt_get_highmem(Object *obj, Error **errp)
{
@@ -2304,6 +2348,7 @@ static void virt_set_acpi(Object *obj, Visitor *v, const char *name,
visit_type_OnOffAuto(v, name, &vms->acpi, errp);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static bool virt_get_ras(Object *obj, Error **errp)
{
VirtMachineState *vms = VIRT_MACHINE(obj);
@@ -2331,6 +2376,7 @@ static void virt_set_mte(Object *obj, bool value, Error **errp)
vms->mte = value;
}
+#endif /* disabled for RHEL */
static char *virt_get_gic_version(Object *obj, Error **errp)
{
@@ -2666,6 +2712,7 @@ static int virt_kvm_type(MachineState *ms, const char *type_str)
return fixed_ipa ? 0 : requested_pa_size;
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void virt_machine_class_init(ObjectClass *oc, void *data)
{
MachineClass *mc = MACHINE_CLASS(oc);
@@ -3031,3 +3078,180 @@ static void virt_machine_2_6_options(MachineClass *mc)
vmc->no_pmu = true;
}
DEFINE_VIRT_MACHINE(2, 6)
+#endif /* disabled for RHEL */
+
+static void rhel_machine_class_init(ObjectClass *oc, void *data)
+{
+ MachineClass *mc = MACHINE_CLASS(oc);
+ HotplugHandlerClass *hc = HOTPLUG_HANDLER_CLASS(oc);
+
+ mc->family = "virt-rhel-Z";
+ mc->init = machvirt_init;
+ /* Maximum supported VCPU count for all virt-rhel* machines */
+ mc->max_cpus = 384;
+#ifdef CONFIG_TPM
+ machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS);
+#endif
+ mc->block_default_type = IF_VIRTIO;
+ mc->no_cdrom = 1;
+ mc->pci_allow_0_address = true;
+ /* We know we will never create a pre-ARMv7 CPU which needs 1K pages */
+ mc->minimum_page_bits = 12;
+ mc->possible_cpu_arch_ids = virt_possible_cpu_arch_ids;
+ mc->cpu_index_to_instance_props = virt_cpu_index_to_props;
+ mc->default_cpu_type = ARM_CPU_TYPE_NAME("cortex-a57");
+ mc->get_default_cpu_node_id = virt_get_default_cpu_node_id;
+ mc->kvm_type = virt_kvm_type;
+ assert(!mc->get_hotplug_handler);
+ mc->get_hotplug_handler = virt_machine_get_hotplug_handler;
+ hc->pre_plug = virt_machine_device_pre_plug_cb;
+ hc->plug = virt_machine_device_plug_cb;
+ hc->unplug_request = virt_machine_device_unplug_request_cb;
+ hc->unplug = virt_machine_device_unplug_cb;
+ mc->nvdimm_supported = true;
+ mc->auto_enable_numa_with_memhp = true;
+ mc->auto_enable_numa_with_memdev = true;
+ mc->default_ram_id = "mach-virt.ram";
+
+ object_class_property_add(oc, "acpi", "OnOffAuto",
+ virt_get_acpi, virt_set_acpi,
+ NULL, NULL);
+ object_class_property_set_description(oc, "acpi",
+ "Enable ACPI");
+
+ object_class_property_add_bool(oc, "highmem", virt_get_highmem,
+ virt_set_highmem);
+ object_class_property_set_description(oc, "highmem",
+ "Set on/off to enable/disable using "
+ "physical address space above 32 bits");
+
+ object_class_property_add_str(oc, "gic-version", virt_get_gic_version,
+ virt_set_gic_version);
+ object_class_property_set_description(oc, "gic-version",
+ "Set GIC version. "
+ "Valid values are 2, 3, host and max");
+
+ object_class_property_add_str(oc, "x-oem-id",
+ virt_get_oem_id,
+ virt_set_oem_id);
+ object_class_property_set_description(oc, "x-oem-id",
+ "Override the default value of field OEMID "
+ "in ACPI table header."
+ "The string may be up to 6 bytes in size");
+
+ object_class_property_add_str(oc, "x-oem-table-id",
+ virt_get_oem_table_id,
+ virt_set_oem_table_id);
+ object_class_property_set_description(oc, "x-oem-table-id",
+ "Override the default value of field OEM Table ID "
+ "in ACPI table header."
+ "The string may be up to 8 bytes in size");
+ object_class_property_add_bool(oc, "default_bus_bypass_iommu",
+ virt_get_default_bus_bypass_iommu,
+ virt_set_default_bus_bypass_iommu);
+
+}
+
+static void rhel_virt_instance_init(Object *obj)
+{
+ VirtMachineState *vms = VIRT_MACHINE(obj);
+ VirtMachineClass *vmc = VIRT_MACHINE_GET_CLASS(vms);
+
+ /* EL3 is disabled by default and non-configurable for RHEL */
+ vms->secure = false;
+
+ /* EL2 is disabled by default and non-configurable for RHEL */
+ vms->virt = false;
+
+ /* High memory is enabled by default */
+ vms->highmem = true;
+ vms->gic_version = VIRT_GIC_VERSION_NOSEL;
+
+ vms->highmem_ecam = !vmc->no_highmem_ecam;
+
+ if (vmc->no_its) {
+ vms->its = false;
+ } else {
+ /* Default allows ITS instantiation */
+ vms->its = true;
+ object_property_add_bool(obj, "its", virt_get_its,
+ virt_set_its);
+ object_property_set_description(obj, "its",
+ "Set on/off to enable/disable "
+ "ITS instantiation");
+ }
+
+ /* Default disallows iommu instantiation */
+ vms->iommu = VIRT_IOMMU_NONE;
+ object_property_add_str(obj, "iommu", virt_get_iommu, virt_set_iommu);
+ object_property_set_description(obj, "iommu",
+ "Set the IOMMU type. "
+ "Valid values are none and smmuv3");
+
+ /* Default disallows RAS instantiation and is non-configurable for RHEL */
+ vms->ras = false;
+
+ /* MTE is disabled by default and non-configurable for RHEL */
+ vms->mte = false;
+
+ vms->default_bus_bypass_iommu = false;
+ vms->irqmap = a15irqmap;
+
+ virt_flash_create(vms);
+ vms->oem_id = g_strndup(ACPI_BUILD_APPNAME6, 6);
+ vms->oem_table_id = g_strndup(ACPI_BUILD_APPNAME8, 8);
+
+}
+
+static const TypeInfo rhel_machine_info = {
+ .name = TYPE_RHEL_MACHINE,
+ .parent = TYPE_MACHINE,
+ .abstract = true,
+ .instance_size = sizeof(VirtMachineState),
+ .class_size = sizeof(VirtMachineClass),
+ .class_init = rhel_machine_class_init,
+ .instance_init = rhel_virt_instance_init,
+ .interfaces = (InterfaceInfo[]) {
+ { TYPE_HOTPLUG_HANDLER },
+ { }
+ },
+};
+
+static void rhel_machine_init(void)
+{
+ type_register_static(&rhel_machine_info);
+}
+type_init(rhel_machine_init);
+
+static void rhel850_virt_options(MachineClass *mc)
+{
+ compat_props_add(mc->compat_props, arm_rhel_compat, arm_rhel_compat_len);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_5, hw_compat_rhel_8_5_len);
+}
+DEFINE_RHEL_MACHINE_AS_LATEST(8, 5, 0)
+
+static void rhel840_virt_options(MachineClass *mc)
+{
+ rhel850_virt_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_4, hw_compat_rhel_8_4_len);
+}
+DEFINE_RHEL_MACHINE(8, 4, 0)
+
+static void rhel830_virt_options(MachineClass *mc)
+{
+ VirtMachineClass *vmc = VIRT_MACHINE_CLASS(OBJECT_CLASS(mc));
+
+ rhel840_virt_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_3, hw_compat_rhel_8_3_len);
+ vmc->no_kvm_steal_time = true;
+}
+DEFINE_RHEL_MACHINE(8, 3, 0)
+
+static void rhel820_virt_options(MachineClass *mc)
+{
+ rhel830_virt_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_2, hw_compat_rhel_8_2_len);
+ mc->numa_mem_supported = true;
+ mc->auto_enable_numa_with_memdev = false;
+}
+DEFINE_RHEL_MACHINE(8, 2, 0)
diff --git a/hw/core/machine.c b/hw/core/machine.c
index be4f9864cd..62febde5aa 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -87,6 +87,8 @@ GlobalProperty hw_compat_rhel_8_3[] = {
{ "nvme", "use-intel-id", "on"},
/* hw_compat_rhel_8_3 from hw_compat_5_1 */
{ "pvpanic", "events", "1"}, /* PVPANIC_PANICKED */
+ /* hw_compat_rhel_8_3 from hw_compat_5_1 */
+ { "pl011", "migrate-clk", "off" },
/* hw_compat_rhel_8_3 bz 1912846 */
{ "pci-xhci", "x-rh-late-msi-cap", "off" },
/* hw_compat_rhel_8_3 from hw_compat_5_1 */
diff --git a/include/hw/arm/virt.h b/include/hw/arm/virt.h
index dc6b66ffc8..9364628847 100644
--- a/include/hw/arm/virt.h
+++ b/include/hw/arm/virt.h
@@ -175,9 +175,17 @@ struct VirtMachineState {
#define VIRT_ECAM_ID(high) (high ? VIRT_HIGH_PCIE_ECAM : VIRT_PCIE_ECAM)
+#if 0 /* disabled for Red Hat Enterprise Linux */
#define TYPE_VIRT_MACHINE MACHINE_TYPE_NAME("virt")
OBJECT_DECLARE_TYPE(VirtMachineState, VirtMachineClass, VIRT_MACHINE)
+#else
+#define TYPE_RHEL_MACHINE MACHINE_TYPE_NAME("virt-rhel")
+typedef struct VirtMachineClass VirtMachineClass;
+typedef struct VirtMachineState VirtMachineState;
+DECLARE_OBJ_CHECKERS(VirtMachineState, VirtMachineClass, VIRT_MACHINE, TYPE_RHEL_MACHINE)
+#endif
+
void virt_acpi_setup(VirtMachineState *vms);
bool virt_is_acpi_enabled(VirtMachineState *vms);
--
2.27.0

714
SOURCES/0009-Add-ppc64-machine-types.patch
Unescape View File

@ -0,0 +1,714 @@
From 3c65320ce5b8ad3bb8c0d8fd13a88c464d5c5845 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 13:27:13 +0200
Subject: Add ppc64 machine types
Adding changes to add RHEL machine types for ppc64 architecture.
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
Rebase changes (4.0.0):
- remove instance options and use upstream solution
- Use upstream compat handling
- Replace SPAPR_PCI_2_7_MMIO_WIN_SIZE with value (changed upstream)
- re-add handling of instance_options (removed upstream)
- Use p8 as default for rhel machine types (p9 default upstream)
- sPAPRMachineClass renamed to SpaprMachineClass (upstream)
Rebase changes (4.1.0):
- Update format for compat structures
Rebase notes (weekly-210303):
- Use rhel-8.4.0 hw compat
Merged patches (4.0.0):
- 467d59a redhat: define pseries-rhel8.0.0 machine type
Merged patches (4.1.0):
- f21757edc target/ppc/spapr: Enable mitigations by default for pseries-4.0 machine type
- 2511c63 redhat: sync pseries-rhel7.6.0 with rhel-av-8.0.1
- 89f01da redhat: define pseries-rhel8.1.0 machine type
Merged patches (4.2.0):
- bcba728 redhat: update pseries-rhel8.1.0 machine type
- redhat: update pseries-rhel-7.6.0 machine type (patch 93039)
- redhat: define pseries-rhel8.2.0 machine type (patch 93041)
Merged patches (5.1.0):
- eb121ff spapr: Enable DD2.3 accelerated count cache flush in pseries-5.0 machine (partial)
Merged patches (5.2.0 rc0):
- 311a20f redhat: define pseries-rhel8.3.0 machine type
- 1284167 ppc: Set correct max_cpus value on spapr-rhel* machine types
- 1ab8783 redhat: update pseries-rhel8.2.0 machine type
- b162af531a target/ppc: Add experimental option for enabling secure guests
Merged patches (weekly-201216):
- 943c936df3 redhat: Add spapr_machine_rhel_default_class_options()
- 030b5e6fba redhat: Define pseries-rhel8.4.0 machine type
Merged patches (weekly-210602):
- b7128d8ef7 redhat: Define pseries-rhel8.5.0 machine type
Merged patches (weekly-211006):
- c8f68b47e9 redhat: Update pseries-rhel8.5.0
---
hw/ppc/spapr.c | 382 ++++++++++++++++++++++++++++++++++++++++
hw/ppc/spapr_cpu_core.c | 13 ++
include/hw/ppc/spapr.h | 4 +
target/ppc/compat.c | 13 +-
target/ppc/cpu.h | 1 +
target/ppc/kvm.c | 27 +++
target/ppc/kvm_ppc.h | 13 ++
7 files changed, 452 insertions(+), 1 deletion(-)
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index 3b5fd749be..cace86028d 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -1593,6 +1593,9 @@ static void spapr_machine_reset(MachineState *machine)
pef_kvm_reset(machine->cgs, &error_fatal);
spapr_caps_apply(spapr);
+ if (spapr->svm_allowed) {
+ kvmppc_svm_allow(&error_fatal);
+ }
first_ppc_cpu = POWERPC_CPU(first_cpu);
if (kvm_enabled() && kvmppc_has_cap_mmu_radix() &&
@@ -3288,6 +3291,20 @@ static void spapr_set_host_serial(Object *obj, const char *value, Error **errp)
spapr->host_serial = g_strdup(value);
}
+static bool spapr_get_svm_allowed(Object *obj, Error **errp)
+{
+ SpaprMachineState *spapr = SPAPR_MACHINE(obj);
+
+ return spapr->svm_allowed;
+}
+
+static void spapr_set_svm_allowed(Object *obj, bool value, Error **errp)
+{
+ SpaprMachineState *spapr = SPAPR_MACHINE(obj);
+
+ spapr->svm_allowed = value;
+}
+
static void spapr_instance_init(Object *obj)
{
SpaprMachineState *spapr = SPAPR_MACHINE(obj);
@@ -3366,6 +3383,12 @@ static void spapr_instance_init(Object *obj)
spapr_get_host_serial, spapr_set_host_serial);
object_property_set_description(obj, "host-serial",
"Host serial number to advertise in guest device tree");
+ object_property_add_bool(obj, "x-svm-allowed",
+ spapr_get_svm_allowed,
+ spapr_set_svm_allowed);
+ object_property_set_description(obj, "x-svm-allowed",
+ "Allow the guest to become a Secure Guest"
+ " (experimental only)");
}
static void spapr_machine_finalizefn(Object *obj)
@@ -4614,6 +4637,7 @@ static void spapr_machine_class_init(ObjectClass *oc, void *data)
vmc->client_architecture_support = spapr_vof_client_architecture_support;
vmc->quiesce = spapr_vof_quiesce;
vmc->setprop = spapr_vof_setprop;
+ smc->has_power9_support = true;
}
static const TypeInfo spapr_machine_info = {
@@ -4665,6 +4689,7 @@ static void spapr_machine_latest_class_options(MachineClass *mc)
} \
type_init(spapr_machine_register_##suffix)
+#if 0 /* Disabled for Red Hat Enterprise Linux */
/*
* pseries-6.2
*/
@@ -4781,6 +4806,7 @@ static void spapr_machine_4_1_class_options(MachineClass *mc)
}
DEFINE_SPAPR_MACHINE(4_1, "4.1", false);
+#endif
/*
* pseries-4.0
@@ -4800,6 +4826,8 @@ static bool phb_placement_4_0(SpaprMachineState *spapr, uint32_t index,
*nv2atsd = 0;
return true;
}
+
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void spapr_machine_4_0_class_options(MachineClass *mc)
{
SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
@@ -4958,6 +4986,7 @@ DEFINE_SPAPR_MACHINE(2_8, "2.8", false);
/*
* pseries-2.7
*/
+#endif
static bool phb_placement_2_7(SpaprMachineState *spapr, uint32_t index,
uint64_t *buid, hwaddr *pio,
@@ -5013,6 +5042,7 @@ static bool phb_placement_2_7(SpaprMachineState *spapr, uint32_t index,
return true;
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void spapr_machine_2_7_class_options(MachineClass *mc)
{
SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
@@ -5127,6 +5157,358 @@ static void spapr_machine_2_1_class_options(MachineClass *mc)
compat_props_add(mc->compat_props, hw_compat_2_1, hw_compat_2_1_len);
}
DEFINE_SPAPR_MACHINE(2_1, "2.1", false);
+#endif
+
+static void spapr_machine_rhel_default_class_options(MachineClass *mc)
+{
+ /*
+ * Defaults for the latest behaviour inherited from the base class
+ * can be overriden here for all pseries-rhel* machines.
+ */
+
+ /* Maximum supported VCPU count */
+ mc->max_cpus = 384;
+}
+
+/*
+ * pseries-rhel8.5.0
+ * like pseries-6.0
+ */
+
+static void spapr_machine_rhel850_class_options(MachineClass *mc)
+{
+ /* The default machine type must apply the RHEL specific defaults */
+ spapr_machine_rhel_default_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_5,
+ hw_compat_rhel_8_5_len);
+}
+
+DEFINE_SPAPR_MACHINE(rhel850, "rhel8.5.0", true);
+
+/*
+ * pseries-rhel8.4.0
+ * like pseries-5.2
+ */
+
+static void spapr_machine_rhel840_class_options(MachineClass *mc)
+{
+ spapr_machine_rhel850_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_4,
+ hw_compat_rhel_8_4_len);
+}
+
+DEFINE_SPAPR_MACHINE(rhel840, "rhel8.4.0", false);
+
+/*
+ * pseries-rhel8.3.0
+ * like pseries-5.1
+ */
+
+static void spapr_machine_rhel830_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel840_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_3,
+ hw_compat_rhel_8_3_len);
+
+ /* from pseries-5.1 */
+ smc->pre_5_2_numa_associativity = true;
+}
+
+DEFINE_SPAPR_MACHINE(rhel830, "rhel8.3.0", false);
+
+/*
+ * pseries-rhel8.2.0
+ * like pseries-4.2 + pseries-5.0
+ * except SPAPR_CAP_CCF_ASSIST that has been backported to pseries-rhel8.1.0
+ */
+
+static void spapr_machine_rhel820_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+ /* from pseries-5.0 */
+ static GlobalProperty compat[] = {
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "pre-5.1-associativity", "on" },
+ };
+
+ spapr_machine_rhel830_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_2,
+ hw_compat_rhel_8_2_len);
+ compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
+
+ /* from pseries-4.2 */
+ smc->default_caps.caps[SPAPR_CAP_FWNMI] = SPAPR_CAP_OFF;
+ smc->rma_limit = 16 * GiB;
+ mc->nvdimm_supported = false;
+
+ /* from pseries-5.0 */
+ mc->numa_mem_supported = true;
+ smc->pre_5_1_assoc_refpoints = true;
+}
+
+DEFINE_SPAPR_MACHINE(rhel820, "rhel8.2.0", false);
+
+/*
+ * pseries-rhel8.1.0
+ * like pseries-4.1
+ */
+
+static void spapr_machine_rhel810_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+ static GlobalProperty compat[] = {
+ /* Only allow 4kiB and 64kiB IOMMU pagesizes */
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "pgsz", "0x11000" },
+ };
+
+ spapr_machine_rhel820_class_options(mc);
+
+ /* from pseries-4.1 */
+ smc->linux_pci_probe = false;
+ smc->smp_threads_vsmt = false;
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_1,
+ hw_compat_rhel_8_1_len);
+ compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
+
+ /* from pseries-4.2 */
+ smc->default_caps.caps[SPAPR_CAP_CCF_ASSIST] = SPAPR_CAP_OFF;
+}
+
+DEFINE_SPAPR_MACHINE(rhel810, "rhel8.1.0", false);
+
+/*
+ * pseries-rhel8.0.0
+ * like pseries-3.1 and pseries-4.0
+ * except SPAPR_CAP_CFPC, SPAPR_CAP_SBBC and SPAPR_CAP_IBS
+ * that have been backported to pseries-rhel8.0.0
+ */
+
+static void spapr_machine_rhel800_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel810_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_0,
+ hw_compat_rhel_8_0_len);
+
+ /* pseries-4.0 */
+ smc->phb_placement = phb_placement_4_0;
+ smc->irq = &spapr_irq_xics;
+ smc->pre_4_1_migration = true;
+
+ /* pseries-3.1 */
+ mc->default_cpu_type = POWERPC_CPU_TYPE_NAME("power8_v2.0");
+ smc->update_dt_enabled = false;
+ smc->dr_phb_enabled = false;
+ smc->broken_host_serial_model = true;
+ smc->default_caps.caps[SPAPR_CAP_LARGE_DECREMENTER] = SPAPR_CAP_OFF;
+}
+
+DEFINE_SPAPR_MACHINE(rhel800, "rhel8.0.0", false);
+
+/*
+ * pseries-rhel7.6.0
+ * like spapr_compat_2_12 and spapr_compat_3_0
+ * spapr_compat_0 is empty
+ */
+GlobalProperty spapr_compat_rhel7_6[] = {
+ { TYPE_POWERPC_CPU, "pre-3.0-migration", "on" },
+ { TYPE_SPAPR_CPU_CORE, "pre-3.0-migration", "on" },
+};
+const size_t spapr_compat_rhel7_6_len = G_N_ELEMENTS(spapr_compat_rhel7_6);
+
+
+static void spapr_machine_rhel760_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel800_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_6, hw_compat_rhel_7_6_len);
+ compat_props_add(mc->compat_props, spapr_compat_rhel7_6, spapr_compat_rhel7_6_len);
+
+ /* from spapr_machine_3_0_class_options() */
+ smc->legacy_irq_allocation = true;
+ smc->nr_xirqs = 0x400;
+ smc->irq = &spapr_irq_xics_legacy;
+
+ /* from spapr_machine_2_12_class_options() */
+ /* We depend on kvm_enabled() to choose a default value for the
+ * hpt-max-page-size capability. Of course we can't do it here
+ * because this is too early and the HW accelerator isn't initialzed
+ * yet. Postpone this to machine init (see default_caps_with_cpu()).
+ */
+ smc->default_caps.caps[SPAPR_CAP_HPT_MAXPAGESIZE] = 0;
+
+ /* SPAPR_CAP_WORKAROUND enabled in pseries-rhel800 by
+ * f21757edc554
+ * "Enable mitigations by default for pseries-4.0 machine type")
+ */
+ smc->default_caps.caps[SPAPR_CAP_CFPC] = SPAPR_CAP_BROKEN;
+ smc->default_caps.caps[SPAPR_CAP_SBBC] = SPAPR_CAP_BROKEN;
+ smc->default_caps.caps[SPAPR_CAP_IBS] = SPAPR_CAP_BROKEN;
+}
+
+DEFINE_SPAPR_MACHINE(rhel760, "rhel7.6.0", false);
+
+/*
+ * pseries-rhel7.6.0-sxxm
+ *
+ * pseries-rhel7.6.0 with speculative execution exploit mitigations enabled by default
+ */
+
+static void spapr_machine_rhel760sxxm_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel760_class_options(mc);
+ smc->default_caps.caps[SPAPR_CAP_CFPC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_SBBC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_IBS] = SPAPR_CAP_FIXED_CCD;
+}
+
+DEFINE_SPAPR_MACHINE(rhel760sxxm, "rhel7.6.0-sxxm", false);
+
+static void spapr_machine_rhel750_class_options(MachineClass *mc)
+{
+ spapr_machine_rhel760_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_5, hw_compat_rhel_7_5_len);
+
+}
+
+DEFINE_SPAPR_MACHINE(rhel750, "rhel7.5.0", false);
+
+/*
+ * pseries-rhel7.5.0-sxxm
+ *
+ * pseries-rhel7.5.0 with speculative execution exploit mitigations enabled by default
+ */
+
+static void spapr_machine_rhel750sxxm_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel750_class_options(mc);
+ smc->default_caps.caps[SPAPR_CAP_CFPC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_SBBC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_IBS] = SPAPR_CAP_FIXED_CCD;
+}
+
+DEFINE_SPAPR_MACHINE(rhel750sxxm, "rhel7.5.0-sxxm", false);
+
+/*
+ * pseries-rhel7.4.0
+ * like spapr_compat_2_9
+ */
+GlobalProperty spapr_compat_rhel7_4[] = {
+ { TYPE_POWERPC_CPU, "pre-2.10-migration", "on" },
+};
+const size_t spapr_compat_rhel7_4_len = G_N_ELEMENTS(spapr_compat_rhel7_4);
+
+static void spapr_machine_rhel740_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel750_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_4, hw_compat_rhel_7_4_len);
+ compat_props_add(mc->compat_props, spapr_compat_rhel7_4, spapr_compat_rhel7_4_len);
+ smc->has_power9_support = false;
+ smc->pre_2_10_has_unused_icps = true;
+ smc->resize_hpt_default = SPAPR_RESIZE_HPT_DISABLED;
+ smc->default_caps.caps[SPAPR_CAP_HTM] = SPAPR_CAP_ON;
+}
+
+DEFINE_SPAPR_MACHINE(rhel740, "rhel7.4.0", false);
+
+/*
+ * pseries-rhel7.4.0-sxxm
+ *
+ * pseries-rhel7.4.0 with speculative execution exploit mitigations enabled by default
+ */
+
+static void spapr_machine_rhel740sxxm_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel740_class_options(mc);
+ smc->default_caps.caps[SPAPR_CAP_CFPC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_SBBC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_IBS] = SPAPR_CAP_FIXED_CCD;
+}
+
+DEFINE_SPAPR_MACHINE(rhel740sxxm, "rhel7.4.0-sxxm", false);
+
+/*
+ * pseries-rhel7.3.0
+ * like spapr_compat_2_6/_2_7/_2_8 but "ddw" has been backported to RHEL7_3
+ */
+GlobalProperty spapr_compat_rhel7_3[] = {
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "mem_win_size", "0xf80000000" },
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "mem64_win_size", "0" },
+ { TYPE_POWERPC_CPU, "pre-2.8-migration", "on" },
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "pre-2.8-migration", "on" },
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "pcie-extended-configuration-space", "off" },
+};
+const size_t spapr_compat_rhel7_3_len = G_N_ELEMENTS(spapr_compat_rhel7_3);
+
+static void spapr_machine_rhel730_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel740_class_options(mc);
+ mc->default_cpu_type = POWERPC_CPU_TYPE_NAME("power7_v2.3");
+ mc->default_machine_opts = "modern-hotplug-events=off";
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_3, hw_compat_rhel_7_3_len);
+ compat_props_add(mc->compat_props, spapr_compat_rhel7_3, spapr_compat_rhel7_3_len);
+
+ smc->phb_placement = phb_placement_2_7;
+}
+
+DEFINE_SPAPR_MACHINE(rhel730, "rhel7.3.0", false);
+
+/*
+ * pseries-rhel7.3.0-sxxm
+ *
+ * pseries-rhel7.3.0 with speculative execution exploit mitigations enabled by default
+ */
+
+static void spapr_machine_rhel730sxxm_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel730_class_options(mc);
+ smc->default_caps.caps[SPAPR_CAP_CFPC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_SBBC] = SPAPR_CAP_WORKAROUND;
+ smc->default_caps.caps[SPAPR_CAP_IBS] = SPAPR_CAP_FIXED_CCD;
+}
+
+DEFINE_SPAPR_MACHINE(rhel730sxxm, "rhel7.3.0-sxxm", false);
+
+/*
+ * pseries-rhel7.2.0
+ */
+/* Should be like spapr_compat_2_5 + 2_4 + 2_3, but "dynamic-reconfiguration"
+ * has been backported to RHEL7_2 so we don't need it here.
+ */
+
+GlobalProperty spapr_compat_rhel7_2[] = {
+ { "spapr-vlan", "use-rx-buffer-pools", "off" },
+ { TYPE_SPAPR_PCI_HOST_BRIDGE, "ddw", "off" },
+};
+const size_t spapr_compat_rhel7_2_len = G_N_ELEMENTS(spapr_compat_rhel7_2);
+
+static void spapr_machine_rhel720_class_options(MachineClass *mc)
+{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
+ spapr_machine_rhel730_class_options(mc);
+ smc->use_ohci_by_default = true;
+ mc->has_hotpluggable_cpus = NULL;
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_2, hw_compat_rhel_7_2_len);
+ compat_props_add(mc->compat_props, spapr_compat_rhel7_2, spapr_compat_rhel7_2_len);
+}
+
+DEFINE_SPAPR_MACHINE(rhel720, "rhel7.2.0", false);
static void spapr_machine_register_types(void)
{
diff --git a/hw/ppc/spapr_cpu_core.c b/hw/ppc/spapr_cpu_core.c
index 8ba34f6a1d..78eca1c04a 100644
--- a/hw/ppc/spapr_cpu_core.c
+++ b/hw/ppc/spapr_cpu_core.c
@@ -24,6 +24,7 @@
#include "sysemu/reset.h"
#include "sysemu/hw_accel.h"
#include "qemu/error-report.h"
+#include "cpu-models.h"
static void spapr_reset_vcpu(PowerPCCPU *cpu)
{
@@ -250,6 +251,7 @@ static bool spapr_realize_vcpu(PowerPCCPU *cpu, SpaprMachineState *spapr,
{
CPUPPCState *env = &cpu->env;
CPUState *cs = CPU(cpu);
+ SpaprMachineClass *smc = SPAPR_MACHINE_GET_CLASS(spapr);
if (!qdev_realize(DEVICE(cpu), NULL, errp)) {
return false;
@@ -261,6 +263,17 @@ static bool spapr_realize_vcpu(PowerPCCPU *cpu, SpaprMachineState *spapr,
cpu_ppc_set_vhyp(cpu, PPC_VIRTUAL_HYPERVISOR(spapr));
kvmppc_set_papr(cpu);
+ if (!smc->has_power9_support &&
+ (((spapr->max_compat_pvr &&
+ ppc_compat_cmp(spapr->max_compat_pvr,
+ CPU_POWERPC_LOGICAL_3_00) >= 0)) ||
+ (!spapr->max_compat_pvr &&
+ ppc_check_compat(cpu, CPU_POWERPC_LOGICAL_3_00, 0, 0)))) {
+ error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
+ "POWER9 CPU is not supported by this machine class");
+ return false;
+ }
+
if (spapr_irq_cpu_intc_create(spapr, cpu, errp) < 0) {
qdev_unrealize(DEVICE(cpu));
return false;
diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h
index ee7504b976..37a014d59c 100644
--- a/include/hw/ppc/spapr.h
+++ b/include/hw/ppc/spapr.h
@@ -154,6 +154,7 @@ struct SpaprMachineClass {
bool pre_5_2_numa_associativity;
bool pre_6_2_numa_affinity;
+ bool has_power9_support;
bool (*phb_placement)(SpaprMachineState *spapr, uint32_t index,
uint64_t *buid, hwaddr *pio,
hwaddr *mmio32, hwaddr *mmio64,
@@ -237,6 +238,9 @@ struct SpaprMachineState {
/* Set by -boot */
char *boot_device;
+
+ /* Secure Guest support via x-svm-allowed */
+ bool svm_allowed;
/*< public >*/
char *kvm_type;
diff --git a/target/ppc/compat.c b/target/ppc/compat.c
index 7949a24f5a..f207a9ba01 100644
--- a/target/ppc/compat.c
+++ b/target/ppc/compat.c
@@ -114,8 +114,19 @@ static const CompatInfo *compat_by_pvr(uint32_t pvr)
return NULL;
}
+long ppc_compat_cmp(uint32_t pvr1, uint32_t pvr2)
+{
+ const CompatInfo *compat1 = compat_by_pvr(pvr1);
+ const CompatInfo *compat2 = compat_by_pvr(pvr2);
+
+ g_assert(compat1);
+ g_assert(compat2);
+
+ return compat1 - compat2;
+}
+
static bool pcc_compat(PowerPCCPUClass *pcc, uint32_t compat_pvr,
- uint32_t min_compat_pvr, uint32_t max_compat_pvr)
+ uint32_t min_compat_pvr, uint32_t max_compat_pvr)
{
const CompatInfo *compat = compat_by_pvr(compat_pvr);
const CompatInfo *min = compat_by_pvr(min_compat_pvr);
diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
index e946da5f3a..23e8b76c85 100644
--- a/target/ppc/cpu.h
+++ b/target/ppc/cpu.h
@@ -1401,6 +1401,7 @@ static inline int cpu_mmu_index(CPUPPCState *env, bool ifetch)
/* Compatibility modes */
#if defined(TARGET_PPC64)
+long ppc_compat_cmp(uint32_t pvr1, uint32_t pvr2);
bool ppc_check_compat(PowerPCCPU *cpu, uint32_t compat_pvr,
uint32_t min_compat_pvr, uint32_t max_compat_pvr);
bool ppc_type_check_compat(const char *cputype, uint32_t compat_pvr,
diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c
index dc93b99189..154888cce5 100644
--- a/target/ppc/kvm.c
+++ b/target/ppc/kvm.c
@@ -90,6 +90,7 @@ static int cap_ppc_nested_kvm_hv;
static int cap_large_decr;
static int cap_fwnmi;
static int cap_rpt_invalidate;
+static int cap_ppc_secure_guest;
static uint32_t debug_inst_opcode;
@@ -137,6 +138,7 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
cap_resize_hpt = kvm_vm_check_extension(s, KVM_CAP_SPAPR_RESIZE_HPT);
kvmppc_get_cpu_characteristics(s);
cap_ppc_nested_kvm_hv = kvm_vm_check_extension(s, KVM_CAP_PPC_NESTED_HV);
+ cap_ppc_secure_guest = kvm_vm_check_extension(s, KVM_CAP_PPC_SECURE_GUEST);
cap_large_decr = kvmppc_get_dec_bits();
cap_fwnmi = kvm_vm_check_extension(s, KVM_CAP_PPC_FWNMI);
/*
@@ -2563,6 +2565,16 @@ int kvmppc_has_cap_rpt_invalidate(void)
return cap_rpt_invalidate;
}
+bool kvmppc_has_cap_secure_guest(void)
+{
+ return !!cap_ppc_secure_guest;
+}
+
+int kvmppc_enable_cap_secure_guest(void)
+{
+ return kvm_vm_enable_cap(kvm_state, KVM_CAP_PPC_SECURE_GUEST, 0, 1);
+}
+
PowerPCCPUClass *kvm_ppc_get_host_cpu_class(void)
{
uint32_t host_pvr = mfpvr();
@@ -2959,3 +2971,18 @@ bool kvm_arch_cpu_check_are_resettable(void)
{
return true;
}
+
+void kvmppc_svm_allow(Error **errp)
+{
+ if (!kvm_enabled()) {
+ error_setg(errp, "No PEF support in tcg, try x-svm-allowed=off");
+ return;
+ }
+
+ if (!kvmppc_has_cap_secure_guest()) {
+ error_setg(errp, "KVM implementation does not support secure guests, "
+ "try x-svm-allowed=off");
+ } else if (kvmppc_enable_cap_secure_guest() < 0) {
+ error_setg(errp, "Error enabling x-svm-allowed, try x-svm-allowed=off");
+ }
+}
diff --git a/target/ppc/kvm_ppc.h b/target/ppc/kvm_ppc.h
index ee9325bf9a..20dbb95989 100644
--- a/target/ppc/kvm_ppc.h
+++ b/target/ppc/kvm_ppc.h
@@ -40,6 +40,7 @@ int kvmppc_booke_watchdog_enable(PowerPCCPU *cpu);
target_ulong kvmppc_configure_v3_mmu(PowerPCCPU *cpu,
bool radix, bool gtse,
uint64_t proc_tbl);
+void kvmppc_svm_allow(Error **errp);
#ifndef CONFIG_USER_ONLY
bool kvmppc_spapr_use_multitce(void);
int kvmppc_spapr_enable_inkernel_multitce(void);
@@ -74,6 +75,8 @@ int kvmppc_get_cap_large_decr(void);
int kvmppc_enable_cap_large_decr(PowerPCCPU *cpu, int enable);
int kvmppc_has_cap_rpt_invalidate(void);
int kvmppc_enable_hwrng(void);
+bool kvmppc_has_cap_secure_guest(void);
+int kvmppc_enable_cap_secure_guest(void);
int kvmppc_put_books_sregs(PowerPCCPU *cpu);
PowerPCCPUClass *kvm_ppc_get_host_cpu_class(void);
void kvmppc_check_papr_resize_hpt(Error **errp);
@@ -393,6 +396,16 @@ static inline int kvmppc_has_cap_rpt_invalidate(void)
return false;
}
+static inline bool kvmppc_has_cap_secure_guest(void)
+{
+ return false;
+}
+
+static inline int kvmppc_enable_cap_secure_guest(void)
+{
+ return -1;
+}
+
static inline int kvmppc_enable_hwrng(void)
{
return -1;
--
2.27.0

165
SOURCES/0010-Add-s390x-machine-types.patch
Unescape View File

@ -0,0 +1,165 @@
From 4ad9a0d0582eef78946b47563eb2c5b7ddf0cbb0 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Fri, 19 Oct 2018 13:47:32 +0200
Subject: Add s390x machine types
Adding changes to add RHEL machine types for s390x architecture.
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
Rebase changes (weekly-4.1.0):
- Use upstream compat handling
Rebase notes (weekly-210303):
- Use rhel-8.4.0 hw compat
Merged patches (3.1.0):
- 29df663 s390x/cpumodel: default enable bpb and ppa15 for z196 and later
Merged patches (4.1.0):
- 6c200d665b hw/s390x/s390-virtio-ccw: Add machine types for RHEL8.0.0
Merged patches (4.2.0):
- fb192e5 redhat: s390x: Rename s390-ccw-virtio-rhel8.0.0 to s390-ccw-virtio-rhel8.1.0
- a9b22e8 redhat: s390x: Add proper compatibility options for the -rhel7.6.0 machine
- hw/s390x: Add the s390-ccw-virtio-rhel8.2.0 machine types (patch 92954)
Merged patches (weekly-201216):
- a6ae745cce redhat: s390x: add rhel-8.4.0 compat machine
Merged patches (weekly-210602):
- 50835d3429 redhat: s390x: add rhel-8.5.0 compat machine
Merged patches (weekly-211006):
- a3bcde27fe redhat: Add s390x machine type compatibility update for 6.1 rebase
---
hw/s390x/s390-virtio-ccw.c | 99 +++++++++++++++++++++++++++++++++++++-
1 file changed, 98 insertions(+), 1 deletion(-)
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index 653587ea62..181856e6cf 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -767,7 +767,7 @@ bool css_migration_enabled(void)
{ \
MachineClass *mc = MACHINE_CLASS(oc); \
ccw_machine_##suffix##_class_options(mc); \
- mc->desc = "VirtIO-ccw based S390 machine v" verstr; \
+ mc->desc = "VirtIO-ccw based S390 machine " verstr; \
if (latest) { \
mc->alias = "s390-ccw-virtio"; \
mc->is_default = true; \
@@ -791,6 +791,7 @@ bool css_migration_enabled(void)
} \
type_init(ccw_machine_register_##suffix)
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void ccw_machine_6_2_instance_options(MachineState *machine)
{
}
@@ -1100,6 +1101,102 @@ static void ccw_machine_2_4_class_options(MachineClass *mc)
compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
}
DEFINE_CCW_MACHINE(2_4, "2.4", false);
+#endif
+
+static void ccw_machine_rhel850_instance_options(MachineState *machine)
+{
+}
+
+static void ccw_machine_rhel850_class_options(MachineClass *mc)
+{
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_5, hw_compat_rhel_8_5_len);
+}
+DEFINE_CCW_MACHINE(rhel850, "rhel8.5.0", true);
+
+static void ccw_machine_rhel840_instance_options(MachineState *machine)
+{
+ ccw_machine_rhel850_instance_options(machine);
+}
+
+static void ccw_machine_rhel840_class_options(MachineClass *mc)
+{
+ ccw_machine_rhel850_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_4, hw_compat_rhel_8_4_len);
+}
+DEFINE_CCW_MACHINE(rhel840, "rhel8.4.0", false);
+
+static void ccw_machine_rhel820_instance_options(MachineState *machine)
+{
+ ccw_machine_rhel840_instance_options(machine);
+}
+
+static void ccw_machine_rhel820_class_options(MachineClass *mc)
+{
+ ccw_machine_rhel840_class_options(mc);
+ mc->fixup_ram_size = s390_fixup_ram_size;
+ /* we did not publish a rhel8.3.0 machine */
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_3, hw_compat_rhel_8_3_len);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_2, hw_compat_rhel_8_2_len);
+}
+DEFINE_CCW_MACHINE(rhel820, "rhel8.2.0", false);
+
+static void ccw_machine_rhel760_instance_options(MachineState *machine)
+{
+ static const S390FeatInit qemu_cpu_feat = { S390_FEAT_LIST_QEMU_V3_1 };
+
+ ccw_machine_rhel820_instance_options(machine);
+
+ s390_set_qemu_cpu_model(0x2827, 12, 2, qemu_cpu_feat);
+
+ /* The multiple-epoch facility was not available with rhel7.6.0 on z14GA1 */
+ s390_cpudef_featoff(14, 1, S390_FEAT_MULTIPLE_EPOCH);
+ s390_cpudef_featoff(14, 1, S390_FEAT_PTFF_QSIE);
+ s390_cpudef_featoff(14, 1, S390_FEAT_PTFF_QTOUE);
+ s390_cpudef_featoff(14, 1, S390_FEAT_PTFF_STOE);
+ s390_cpudef_featoff(14, 1, S390_FEAT_PTFF_STOUE);
+}
+
+static void ccw_machine_rhel760_class_options(MachineClass *mc)
+{
+ ccw_machine_rhel820_class_options(mc);
+ /* We never published the s390x version of RHEL-AV 8.0 and 8.1, so add this here */
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_1, hw_compat_rhel_8_1_len);
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_0, hw_compat_rhel_8_0_len);
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_6, hw_compat_rhel_7_6_len);
+}
+DEFINE_CCW_MACHINE(rhel760, "rhel7.6.0", false);
+
+static void ccw_machine_rhel750_instance_options(MachineState *machine)
+{
+ static const S390FeatInit qemu_cpu_feat = { S390_FEAT_LIST_QEMU_V2_11 };
+ ccw_machine_rhel760_instance_options(machine);
+
+ /* before 2.12 we emulated the very first z900, and RHEL 7.5 is
+ based on 2.10 */
+ s390_set_qemu_cpu_model(0x2064, 7, 1, qemu_cpu_feat);
+
+ /* bpb and ppa15 were only in the full model in RHEL 7.5 */
+ s390_cpudef_featoff_greater(11, 1, S390_FEAT_PPA15);
+ s390_cpudef_featoff_greater(11, 1, S390_FEAT_BPB);
+}
+
+GlobalProperty ccw_compat_rhel_7_5[] = {
+ {
+ .driver = TYPE_SCLP_EVENT_FACILITY,
+ .property = "allow_all_mask_sizes",
+ .value = "off",
+ },
+};
+const size_t ccw_compat_rhel_7_5_len = G_N_ELEMENTS(ccw_compat_rhel_7_5);
+
+static void ccw_machine_rhel750_class_options(MachineClass *mc)
+{
+ ccw_machine_rhel760_class_options(mc);
+ compat_props_add(mc->compat_props, hw_compat_rhel_7_5, hw_compat_rhel_7_5_len);
+ compat_props_add(mc->compat_props, ccw_compat_rhel_7_5, ccw_compat_rhel_7_5_len);
+ S390_CCW_MACHINE_CLASS(mc)->hpage_1m_allowed = false;
+}
+DEFINE_CCW_MACHINE(rhel750, "rhel7.5.0", false);
static void ccw_machine_register_types(void)
{
--
2.27.0

1276
SOURCES/0011-Add-x86_64-machine-types.patch
View File

File diff suppressed because it is too large Load Diff

407
SOURCES/0012-Enable-make-check.patch
Unescape View File

@ -0,0 +1,407 @@
From 740a2dd943a2e0fcd41a9cd8eb94a136f8f49fa2 Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 2 Sep 2020 09:39:41 +0200
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 (4.0.0):
- Remove testing for pseries-2.7 in endianess test
- Disable device-plug-test on s390x as it use disabled device
- Do not run cpu-plug-tests on 7.3 and older machine types
Rebase changes (4.1.0-rc0):
- removed iotests 068
Rebase changes (4.1.0-rc1):
- remove all 205 tests (unstable)
Rebase changes (4.2.0-rc0):
- partially disable hd-geo-test (requires lsi53c895a)
Rebase changes (5.1.0-rc1):
- Disable qtest/q35-test (uses upstream machine types)
- Do not run iotests on make checka
- Enabled iotests 071 and 099
Rebase changes (5.2.0 rc0):
- Disable cdrom tests (unsupported devices) on x86_64
- disable fuzz test
Rebase changes (6.0.0):
- Disabled xlnx-can-test
- Disable pxb-pcie subtest for bios-table-test
- Replace qtest usage of upstream q35 machine type with pc-q35-rhel8.4.0
- Not run cdrom-test on aarch64
Rebase changes (6.1.0):
- Remove unnecessary test disabling changes
Rebase changes (weekly-211006):
- New handling for bios-table-test (disabled downstream)
Merged patches (4.0.0):
- f7ffd13 Remove 7 qcow2 and luks iotests that are taking > 25 sec to run during the fast train build proce
Merged patches (4.1.0-rc0):
- 41288ff redhat: Remove raw iotest 205
---
redhat/qemu-kvm.spec.template | 2 +-
tests/qemu-iotests/051 | 8 ++++----
tests/qtest/bios-tables-test.c | 5 ++++-
tests/qtest/boot-serial-test.c | 6 +++++-
tests/qtest/cdrom-test.c | 4 ++++
tests/qtest/cpu-plug-test.c | 4 ++--
tests/qtest/fuzz-e1000e-test.c | 2 +-
tests/qtest/fuzz-virtio-scsi-test.c | 2 +-
tests/qtest/hd-geo-test.c | 4 ++++
tests/qtest/lpc-ich9-test.c | 2 +-
tests/qtest/meson.build | 13 ++++---------
tests/qtest/prom-env-test.c | 4 ++++
tests/qtest/test-x86-cpuid-compat.c | 2 ++
tests/qtest/usb-hcd-xhci-test.c | 4 ++++
14 files changed, 41 insertions(+), 21 deletions(-)
diff --git a/tests/qemu-iotests/051 b/tests/qemu-iotests/051
index 1d2fa93a11..c8a2815f54 100755
--- a/tests/qemu-iotests/051
+++ b/tests/qemu-iotests/051
@@ -174,9 +174,9 @@ run_qemu -drive if=virtio
case "$QEMU_DEFAULT_MACHINE" in
pc)
run_qemu -drive if=none,id=disk -device ide-cd,drive=disk
- run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-cd,drive=disk
+# run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-cd,drive=disk
run_qemu -drive if=none,id=disk -device ide-hd,drive=disk
- run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-hd,drive=disk
+# run_qemu -drive if=none,id=disk -device lsi53c895a -device scsi-hd,drive=disk
;;
*)
;;
@@ -225,9 +225,9 @@ run_qemu -drive file="$TEST_IMG",if=virtio,readonly=on
case "$QEMU_DEFAULT_MACHINE" in
pc)
run_qemu -drive file="$TEST_IMG",if=none,id=disk,readonly=on -device ide-cd,drive=disk
- run_qemu -drive file="$TEST_IMG",if=none,id=disk,readonly=on -device lsi53c895a -device scsi-cd,drive=disk
+# run_qemu -drive file="$TEST_IMG",if=none,id=disk,readonly=on -device lsi53c895a -device scsi-cd,drive=disk
run_qemu -drive file="$TEST_IMG",if=none,id=disk,readonly=on -device ide-hd,drive=disk
- run_qemu -drive file="$TEST_IMG",if=none,id=disk,readonly=on -device lsi53c895a -device scsi-hd,drive=disk
+# run_qemu -drive file="$TEST_IMG",if=none,id=disk,readonly=on -device lsi53c895a -device scsi-hd,drive=disk
;;
*)
;;
diff --git a/tests/qtest/bios-tables-test.c b/tests/qtest/bios-tables-test.c
index 258874167e..16d8304cde 100644
--- a/tests/qtest/bios-tables-test.c
+++ b/tests/qtest/bios-tables-test.c
@@ -1372,6 +1372,7 @@ static void test_acpi_virt_tcg_numamem(void)
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void test_acpi_virt_tcg_pxb(void)
{
test_data data = {
@@ -1403,6 +1404,7 @@ static void test_acpi_virt_tcg_pxb(void)
free_test_data(&data);
}
+#endif
static void test_acpi_tcg_acpi_hmat(const char *machine)
{
@@ -1644,7 +1646,8 @@ int main(int argc, char *argv[])
qtest_add_func("acpi/virt", test_acpi_virt_tcg);
qtest_add_func("acpi/virt/numamem", test_acpi_virt_tcg_numamem);
qtest_add_func("acpi/virt/memhp", test_acpi_virt_tcg_memhp);
- qtest_add_func("acpi/virt/pxb", test_acpi_virt_tcg_pxb);
+ /* Disabled for Red Hat Enterprise Linux
+ qtest_add_func("acpi/virt/pxb", test_acpi_virt_tcg_pxb); */
qtest_add_func("acpi/virt/oem-fields", test_acpi_oem_fields_virt);
}
}
diff --git a/tests/qtest/boot-serial-test.c b/tests/qtest/boot-serial-test.c
index 83828ba270..294476b959 100644
--- a/tests/qtest/boot-serial-test.c
+++ b/tests/qtest/boot-serial-test.c
@@ -148,19 +148,23 @@ static testdef_t tests[] = {
{ "ppc", "g3beige", "", "PowerPC,750" },
{ "ppc", "mac99", "", "PowerPC,G4" },
{ "ppc", "sam460ex", "-m 256", "DRAM: 256 MiB" },
+#if 0 /* Disabled for Red Hat Enterprise Linux */
{ "ppc64", "ppce500", "", "U-Boot" },
{ "ppc64", "40p", "-m 192", "Memory: 192M" },
{ "ppc64", "mac99", "", "PowerPC,970FX" },
+#endif
{ "ppc64", "pseries",
"-machine " PSERIES_DEFAULT_CAPABILITIES,
"Open Firmware" },
+#if 0 /* Disabled for Red Hat Enterprise Linux */
{ "ppc64", "powernv8", "", "OPAL" },
{ "ppc64", "powernv9", "", "OPAL" },
{ "ppc64", "sam460ex", "-device e1000", "8086 100e" },
+#endif
{ "i386", "isapc", "-cpu qemu32 -device sga", "SGABIOS" },
{ "i386", "pc", "-device sga", "SGABIOS" },
{ "i386", "q35", "-device sga", "SGABIOS" },
- { "x86_64", "isapc", "-cpu qemu32 -device sga", "SGABIOS" },
+ { "x86_64", "pc", "-cpu qemu32 -device sga", "SGABIOS" },
{ "x86_64", "q35", "-device sga", "SGABIOS" },
{ "sparc", "LX", "", "TMS390S10" },
{ "sparc", "SS-4", "", "MB86904" },
diff --git a/tests/qtest/cdrom-test.c b/tests/qtest/cdrom-test.c
index 5af944a5fb..69d9bac38a 100644
--- a/tests/qtest/cdrom-test.c
+++ b/tests/qtest/cdrom-test.c
@@ -140,6 +140,7 @@ static void add_x86_tests(void)
qtest_add_data_func("cdrom/boot/isapc", "-M isapc "
"-drive if=ide,media=cdrom,file=", test_cdboot);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
qtest_add_data_func("cdrom/boot/am53c974",
"-device am53c974 -device scsi-cd,drive=cd1 "
"-drive if=none,id=cd1,format=raw,file=", test_cdboot);
@@ -155,6 +156,7 @@ static void add_x86_tests(void)
qtest_add_data_func("cdrom/boot/megasas-gen2", "-M q35 "
"-device megasas-gen2 -device scsi-cd,drive=cd1 "
"-blockdev file,node-name=cd1,filename=", test_cdboot);
+#endif
}
static void add_s390x_tests(void)
@@ -220,6 +222,7 @@ int main(int argc, char **argv)
"magnum", "malta", "pica61", NULL
};
add_cdrom_param_tests(mips64machines);
+#if 0 /* Disabled for Red Hat Enterprise Linux */
} else if (g_str_equal(arch, "arm") || g_str_equal(arch, "aarch64")) {
const char *armmachines[] = {
"realview-eb", "realview-eb-mpcore", "realview-pb-a8",
@@ -227,6 +230,7 @@ int main(int argc, char **argv)
"vexpress-a9", "virt", NULL
};
add_cdrom_param_tests(armmachines);
+#endif
} else {
const char *nonemachine[] = { "none", NULL };
add_cdrom_param_tests(nonemachine);
diff --git a/tests/qtest/cpu-plug-test.c b/tests/qtest/cpu-plug-test.c
index a1c689414b..a8f076711c 100644
--- a/tests/qtest/cpu-plug-test.c
+++ b/tests/qtest/cpu-plug-test.c
@@ -110,8 +110,8 @@ static void add_pseries_test_case(const char *mname)
char *path;
PlugTestData *data;
- if (!g_str_has_prefix(mname, "pseries-") ||
- (g_str_has_prefix(mname, "pseries-2.") && atoi(&mname[10]) < 7)) {
+ if (!g_str_has_prefix(mname, "pseries-rhel") ||
+ (g_str_has_prefix(mname, "pseries-rhel7.") && atoi(&mname[14]) < 4)) {
return;
}
data = g_new(PlugTestData, 1);
diff --git a/tests/qtest/fuzz-e1000e-test.c b/tests/qtest/fuzz-e1000e-test.c
index 66229e6096..947fba73b7 100644
--- a/tests/qtest/fuzz-e1000e-test.c
+++ b/tests/qtest/fuzz-e1000e-test.c
@@ -17,7 +17,7 @@ static void test_lp1879531_eth_get_rss_ex_dst_addr(void)
{
QTestState *s;
- s = qtest_init("-nographic -monitor none -serial none -M pc-q35-5.0");
+ s = qtest_init("-nographic -monitor none -serial none -M pc-q35-rhel8.4.0");
qtest_outl(s, 0xcf8, 0x80001010);
qtest_outl(s, 0xcfc, 0xe1020000);
diff --git a/tests/qtest/fuzz-virtio-scsi-test.c b/tests/qtest/fuzz-virtio-scsi-test.c
index aaf6d10e18..43727d62ac 100644
--- a/tests/qtest/fuzz-virtio-scsi-test.c
+++ b/tests/qtest/fuzz-virtio-scsi-test.c
@@ -19,7 +19,7 @@ static void test_mmio_oob_from_memory_region_cache(void)
{
QTestState *s;
- s = qtest_init("-M pc-q35-5.2 -display none -m 512M "
+ s = qtest_init("-M pc-q35-rhel8.4.0 -display none -m 512M "
"-device virtio-scsi,num_queues=8,addr=03.0 ");
qtest_outl(s, 0xcf8, 0x80001811);
diff --git a/tests/qtest/hd-geo-test.c b/tests/qtest/hd-geo-test.c
index 113126ae06..999ef2aace 100644
--- a/tests/qtest/hd-geo-test.c
+++ b/tests/qtest/hd-geo-test.c
@@ -737,6 +737,7 @@ static void test_override_ide(void)
test_override(args, expected);
}
+#if 0 /* Require lsi53c895a - not supported on RHEL */
static void test_override_scsi(void)
{
TestArgs *args = create_args();
@@ -781,6 +782,7 @@ static void test_override_scsi_2_controllers(void)
add_scsi_disk(args, 3, 1, 0, 1, 2, 0, 1, 0);
test_override(args, expected);
}
+#endif
static void test_override_virtio_blk(void)
{
@@ -960,9 +962,11 @@ int main(int argc, char **argv)
qtest_add_func("hd-geo/ide/device/user/chst", test_ide_device_user_chst);
if (have_qemu_img()) {
qtest_add_func("hd-geo/override/ide", test_override_ide);
+#if 0 /* Require lsi53c895a - not supported on RHEL */
qtest_add_func("hd-geo/override/scsi", test_override_scsi);
qtest_add_func("hd-geo/override/scsi_2_controllers",
test_override_scsi_2_controllers);
+#endif
qtest_add_func("hd-geo/override/virtio_blk", test_override_virtio_blk);
qtest_add_func("hd-geo/override/zero_chs", test_override_zero_chs);
qtest_add_func("hd-geo/override/scsi_hot_unplug",
diff --git a/tests/qtest/lpc-ich9-test.c b/tests/qtest/lpc-ich9-test.c
index fe0bef9980..7a9d51579b 100644
--- a/tests/qtest/lpc-ich9-test.c
+++ b/tests/qtest/lpc-ich9-test.c
@@ -15,7 +15,7 @@ static void test_lp1878642_pci_bus_get_irq_level_assert(void)
{
QTestState *s;
- s = qtest_init("-M pc-q35-5.0 "
+ s = qtest_init("-M pc-q35-rhel8.4.0 "
"-nographic -monitor none -serial none");
qtest_outl(s, 0xcf8, 0x8000f840); /* PMBASE */
diff --git a/tests/qtest/meson.build b/tests/qtest/meson.build
index c9d8458062..049e06c057 100644
--- a/tests/qtest/meson.build
+++ b/tests/qtest/meson.build
@@ -68,7 +68,6 @@ qtests_i386 = \
(config_all_devices.has_key('CONFIG_RTL8139_PCI') ? ['rtl8139-test'] : []) + \
(config_all_devices.has_key('CONFIG_E1000E_PCI_EXPRESS') ? ['fuzz-e1000e-test'] : []) + \
(config_all_devices.has_key('CONFIG_ESP_PCI') ? ['am53c974-test'] : []) + \
- (unpack_edk2_blobs ? ['bios-tables-test'] : []) + \
qtests_pci + \
['fdc-test',
'ide-test',
@@ -81,7 +80,6 @@ qtests_i386 = \
'drive_del-test',
'tco-test',
'cpu-plug-test',
- 'q35-test',
'vmgenid-test',
'migration-test',
'test-x86-cpuid-compat',
@@ -130,17 +128,15 @@ qtests_mips64el = \
qtests_ppc = \
(config_all_devices.has_key('CONFIG_ISA_TESTDEV') ? ['endianness-test'] : []) + \
- (config_all_devices.has_key('CONFIG_M48T59') ? ['m48t59-test'] : []) + \
- ['boot-order-test', 'prom-env-test', 'boot-serial-test'] \
+ (config_all_devices.has_key('CONFIG_M48T59') ? ['m48t59-test'] : [])
qtests_ppc64 = \
(config_all_devices.has_key('CONFIG_PSERIES') ? ['device-plug-test'] : []) + \
(config_all_devices.has_key('CONFIG_POWERNV') ? ['pnv-xscom-test'] : []) + \
(config_all_devices.has_key('CONFIG_PSERIES') ? ['rtas-test'] : []) + \
- (slirp.found() ? ['pxe-test', 'test-netfilter'] : []) + \
+ (slirp.found() ? ['pxe-test'] : []) + \
(config_all_devices.has_key('CONFIG_USB_UHCI') ? ['usb-hcd-uhci-test'] : []) + \
(config_all_devices.has_key('CONFIG_USB_XHCI_NEC') ? ['usb-hcd-xhci-test'] : []) + \
- (config_host.has_key('CONFIG_POSIX') ? ['test-filter-mirror'] : []) + \
qtests_pci + ['migration-test', 'numa-test', 'cpu-plug-test', 'drive_del-test']
qtests_sh4 = (config_all_devices.has_key('CONFIG_ISA_TESTDEV') ? ['endianness-test'] : [])
@@ -186,8 +182,8 @@ qtests_aarch64 = \
['arm-cpu-features',
'numa-test',
'boot-serial-test',
- 'xlnx-can-test',
- 'fuzz-xlnx-dp-test',
+# 'xlnx-can-test',
+# 'fuzz-xlnx-dp-test',
'migration-test']
qtests_s390x = \
@@ -196,7 +192,6 @@ qtests_s390x = \
(config_host.has_key('CONFIG_POSIX') ? ['test-filter-redirector'] : []) + \
['boot-serial-test',
'drive_del-test',
- 'device-plug-test',
'virtio-ccw-test',
'cpu-plug-test',
'migration-test']
diff --git a/tests/qtest/prom-env-test.c b/tests/qtest/prom-env-test.c
index f41d80154a..f8dc478ce8 100644
--- a/tests/qtest/prom-env-test.c
+++ b/tests/qtest/prom-env-test.c
@@ -89,10 +89,14 @@ int main(int argc, char *argv[])
if (!strcmp(arch, "ppc")) {
add_tests(ppc_machines);
} else if (!strcmp(arch, "ppc64")) {
+#if 0 /* Disabled for Red Hat Enterprise Linux */
add_tests(ppc_machines);
if (g_test_slow()) {
+#endif
qtest_add_data_func("prom-env/pseries", "pseries", test_machine);
+#if 0 /* Disabled for Red Hat Enterprise Linux */
}
+#endif
} else if (!strcmp(arch, "sparc")) {
add_tests(sparc_machines);
} else if (!strcmp(arch, "sparc64")) {
diff --git a/tests/qtest/test-x86-cpuid-compat.c b/tests/qtest/test-x86-cpuid-compat.c
index f28848e06e..6b2fd398a2 100644
--- a/tests/qtest/test-x86-cpuid-compat.c
+++ b/tests/qtest/test-x86-cpuid-compat.c
@@ -300,6 +300,7 @@ int main(int argc, char **argv)
"-cpu 486,xlevel2=0xC0000002,xstore=on",
"xlevel2", 0xC0000002);
+#if 0 /* Disabled in Red Hat Enterprise Linux */
/* Check compatibility of old machine-types that didn't
* auto-increase level/xlevel/xlevel2: */
@@ -350,6 +351,7 @@ int main(int argc, char **argv)
add_cpuid_test("x86/cpuid/xlevel-compat/pc-i440fx-2.4/npt-on",
"-machine pc-i440fx-2.4 -cpu SandyBridge,svm=on,npt=on",
"xlevel", 0x80000008);
+#endif
/* Test feature parsing */
add_feature_test("x86/cpuid/features/plus",
diff --git a/tests/qtest/usb-hcd-xhci-test.c b/tests/qtest/usb-hcd-xhci-test.c
index 10ef9d2a91..3855873050 100644
--- a/tests/qtest/usb-hcd-xhci-test.c
+++ b/tests/qtest/usb-hcd-xhci-test.c
@@ -21,6 +21,7 @@ static void test_xhci_hotplug(void)
usb_test_hotplug(global_qtest, "xhci", "1", NULL);
}
+#if 0 /* Disabled for Red Hat Enterprise Linux */
static void test_usb_uas_hotplug(void)
{
QTestState *qts = global_qtest;
@@ -36,6 +37,7 @@ static void test_usb_uas_hotplug(void)
qtest_qmp_device_del(qts, "scsihd");
qtest_qmp_device_del(qts, "uas");
}
+#endif
static void test_usb_ccid_hotplug(void)
{
@@ -56,7 +58,9 @@ int main(int argc, char **argv)
qtest_add_func("/xhci/pci/init", test_xhci_init);
qtest_add_func("/xhci/pci/hotplug", test_xhci_hotplug);
+#if 0 /* Disabled for Red Hat Enterprise Linux */
qtest_add_func("/xhci/pci/hotplug/usb-uas", test_usb_uas_hotplug);
+#endif
qtest_add_func("/xhci/pci/hotplug/usb-ccid", test_usb_ccid_hotplug);
qtest_start("-device nec-usb-xhci,id=xhci"
--
2.27.0

110
SOURCES/0013-vfio-cap-number-of-devices-that-can-be-assigned.patch
Unescape View File

@ -0,0 +1,110 @@
From e9ebc159a9acf108e1ec6f622be3f256cf14aba7 Mon Sep 17 00:00:00 2001
From: Bandan Das <bsd@redhat.com>
Date: Tue, 3 Dec 2013 20:05:13 +0100
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>
Patchwork-id: 55984
O-Subject: [PATCH RHEL7 qemu-kvm v2 2/2] vfio: cap number of devices that can be assigned
Bugzilla: 678368
RH-Acked-by: Alex Williamson <alex.williamson@redhat.com>
RH-Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
RH-Acked-by: Michael S. Tsirkin <mst@redhat.com>
Go through all groups to get count of total number of devices
active to enforce limit
Reasoning from Alex for the limit(32) - Assuming 3 slots per
device, with 125 slots (number of memory slots for RHEL 7),
we can support almost 40 devices and still have few slots left
for other uses. Stepping down a bit, the number 32 arbitrarily
matches the number of slots on a PCI bus and is also a nice power
of two.
Signed-off-by: Bandan Das <bsd@redhat.com>
Rebase notes (2.8.0):
- removed return value for vfio_realize (commit 1a22aca)
Merged patches (2.9.0):
- 17eb774 vfio: Use error_setg when reporting max assigned device overshoot
Merged patches (4.1.0-rc3):
- 2b89558 vfio: increase the cap on number of assigned devices to 64
---
hw/vfio/pci.c | 29 ++++++++++++++++++++++++++++-
hw/vfio/pci.h | 1 +
2 files changed, 29 insertions(+), 1 deletion(-)
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 7b45353ce2..eb725a3aee 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -45,6 +45,9 @@
#define TYPE_VFIO_PCI_NOHOTPLUG "vfio-pci-nohotplug"
+/* RHEL only: Set once for the first assigned dev */
+static uint16_t device_limit;
+
static void vfio_disable_interrupts(VFIOPCIDevice *vdev);
static void vfio_mmap_set_enabled(VFIOPCIDevice *vdev, bool enabled);
@@ -2807,9 +2810,30 @@ static void vfio_realize(PCIDevice *pdev, Error **errp)
ssize_t len;
struct stat st;
int groupid;
- int i, ret;
+ int ret, i = 0;
bool is_mdev;
+ if (device_limit && device_limit != vdev->assigned_device_limit) {
+ error_setg(errp, "Assigned device limit has been redefined. "
+ "Old:%d, New:%d",
+ device_limit, vdev->assigned_device_limit);
+ return;
+ } else {
+ device_limit = vdev->assigned_device_limit;
+ }
+
+ QLIST_FOREACH(group, &vfio_group_list, next) {
+ QLIST_FOREACH(vbasedev_iter, &group->device_list, next) {
+ i++;
+ }
+ }
+
+ if (i >= vdev->assigned_device_limit) {
+ error_setg(errp, "Maximum supported vfio devices (%d) "
+ "already attached", vdev->assigned_device_limit);
+ return;
+ }
+
if (!vdev->vbasedev.sysfsdev) {
if (!(~vdev->host.domain || ~vdev->host.bus ||
~vdev->host.slot || ~vdev->host.function)) {
@@ -3246,6 +3270,9 @@ static Property vfio_pci_dev_properties[] = {
DEFINE_PROP_BOOL("x-no-kvm-msix", VFIOPCIDevice, no_kvm_msix, false),
DEFINE_PROP_BOOL("x-no-geforce-quirks", VFIOPCIDevice,
no_geforce_quirks, false),
+ /* RHEL only */
+ DEFINE_PROP_UINT16("x-assigned-device-limit", VFIOPCIDevice,
+ assigned_device_limit, 64),
DEFINE_PROP_BOOL("x-no-kvm-ioeventfd", VFIOPCIDevice, no_kvm_ioeventfd,
false),
DEFINE_PROP_BOOL("x-no-vfio-ioeventfd", VFIOPCIDevice, no_vfio_ioeventfd,
diff --git a/hw/vfio/pci.h b/hw/vfio/pci.h
index 64777516d1..e0fe6ca97e 100644
--- a/hw/vfio/pci.h
+++ b/hw/vfio/pci.h
@@ -139,6 +139,7 @@ struct VFIOPCIDevice {
EventNotifier err_notifier;
EventNotifier req_notifier;
int (*resetfn)(struct VFIOPCIDevice *);
+ uint16_t assigned_device_limit;
uint32_t vendor_id;
uint32_t device_id;
uint32_t sub_vendor_id;
--
2.27.0

55
SOURCES/0014-Add-support-statement-to-help-output.patch
Unescape View File

@ -0,0 +1,55 @@
From b736b0c41dd62ed6f874a7b33ca1d4f9ceab4573 Mon Sep 17 00:00:00 2001
From: Eduardo Habkost <ehabkost@redhat.com>
Date: Wed, 4 Dec 2013 18:53:17 +0100
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>
Patchwork-id: 55994
O-Subject: [qemu-kvm RHEL7 PATCH] Add support statement to -help output
Bugzilla: 972773
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: knoel@redhat.com
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Add support statement to -help output, reporting direct qemu-kvm usage
as unsupported by Red Hat, and advising users to use libvirt instead.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
softmmu/vl.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/softmmu/vl.c b/softmmu/vl.c
index 620a1f1367..d46b8fb4ab 100644
--- a/softmmu/vl.c
+++ b/softmmu/vl.c
@@ -827,9 +827,17 @@ static void version(void)
QEMU_COPYRIGHT "\n");
}
+static void print_rh_warning(void)
+{
+ printf("\nWARNING: Direct use of qemu-kvm from the command line is not supported by Red Hat.\n"
+ "WARNING: Use libvirt as the stable management interface.\n"
+ "WARNING: Some command line options listed here may not be available in future releases.\n\n");
+}
+
static void help(int exitcode)
{
version();
+ print_rh_warning();
printf("usage: %s [options] [disk_image]\n\n"
"'disk_image' is a raw hard disk image for IDE hard disk 0\n\n",
error_get_progname());
@@ -855,6 +863,7 @@ static void help(int exitcode)
"\n"
QEMU_HELP_BOTTOM "\n");
+ print_rh_warning();
exit(exitcode);
}
--
2.27.0

65
SOURCES/0015-globally-limit-the-maximum-number-of-CPUs.patch
Unescape View File

@ -0,0 +1,65 @@
From 9a7621819821ee88d2f99d6b629fd87aa9a07758 Mon Sep 17 00:00:00 2001
From: Andrew Jones <drjones@redhat.com>
Date: Tue, 21 Jan 2014 10:46:52 +0100
Subject: globally limit the maximum number of CPUs
We now globally limit the number of VCPUs.
Especially, there is no way one can specify more than
max_cpus VCPUs for a VM.
This allows us the restore the ppc max_cpus limitation to the upstream
default and minimize the ppc hack in kvm-all.c.
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
Signed-off-by: Danilo Cesar Lemes de Paula <ddepaula@redhat.com>
Rebase notes (2.11.0):
- Removed CONFIG_RHV reference
- Update commit log
Merged patches (2.11.0):
- 92fef14623 redhat: remove manual max_cpus limitations for ppc
- bb722e9eff redhat: globally limit the maximum number of CPUs
- fdeef3c1c7 RHEL: Set vcpus hard limit to 240 for Power
- 0584216921 Match POWER max cpus to x86
Signed-off-by: Andrew Jones <drjones@redhat.com>
Merged patches (5.1.0):
- redhat: globally limit the maximum number of CPUs
- redhat: remove manual max_cpus limitations for ppc
- use recommended max vcpu count
Merged patches (5.2.0 rc0):
- f8a4123 vl: Remove downstream-only MAX_RHEL_CPUS code
---
accel/kvm/kvm-all.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index eecd8031cf..8f2a53438f 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2423,6 +2423,18 @@ static int kvm_init(MachineState *ms)
soft_vcpus_limit = kvm_recommended_vcpus(s);
hard_vcpus_limit = kvm_max_vcpus(s);
+#ifdef HOST_PPC64
+ /*
+ * On POWER, the kernel advertises a soft limit based on the
+ * number of CPU threads on the host. We want to allow exceeding
+ * this for testing purposes, so we don't want to set hard limit
+ * to soft limit as on x86.
+ */
+#else
+ /* RHEL doesn't support nr_vcpus > soft_vcpus_limit */
+ hard_vcpus_limit = soft_vcpus_limit;
+#endif
+
while (nc->name) {
if (nc->num > soft_vcpus_limit) {
warn_report("Number of %s cpus requested (%d) exceeds "
--
2.27.0

126
SOURCES/0016-Use-qemu-kvm-in-documentation-instead-of-qemu-system.patch
Unescape View File

@ -0,0 +1,126 @@
From 0d3fc0b4c5773c6cabb0a58c064475f76eb6ac1e Mon Sep 17 00:00:00 2001
From: Miroslav Rezanina <mrezanin@redhat.com>
Date: Wed, 8 Jul 2020 08:35:50 +0200
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
Bugzilla: 1140620
RH-Acked-by: Laszlo Ersek <lersek@redhat.com>
RH-Acked-by: Markus Armbruster <armbru@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
From: Miroslav Rezanina <mrezanin@redhat.com>
We change the name and location of qemu-kvm binaries. Update documentation
to reflect this change. Only architectures available in RHEL are updated.
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
Rebase notes (5.1.0 rc0):
- qemu-block-drivers.texi converted to qemu-block-drivers.rst (upstream)
Rebase notes (5.2.0 rc0):
- rewrite patch to new docs structure
---
docs/defs.rst.inc | 4 ++--
docs/tools/qemu-trace-stap.rst | 14 +++++++-------
qemu-options.hx | 10 +++++-----
3 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/docs/defs.rst.inc b/docs/defs.rst.inc
index 52d6454b93..d74dbdeca9 100644
--- a/docs/defs.rst.inc
+++ b/docs/defs.rst.inc
@@ -9,7 +9,7 @@
but the manpages will end up misrendered with following normal text
incorrectly in boldface.
-.. |qemu_system| replace:: qemu-system-x86_64
-.. |qemu_system_x86| replace:: qemu-system-x86_64
+.. |qemu_system| replace:: qemu-kvm
+.. |qemu_system_x86| replace:: qemu-kvm
.. |I2C| replace:: I\ :sup:`2`\ C
.. |I2S| replace:: I\ :sup:`2`\ S
diff --git a/docs/tools/qemu-trace-stap.rst b/docs/tools/qemu-trace-stap.rst
index d53073b52b..9e93df084f 100644
--- a/docs/tools/qemu-trace-stap.rst
+++ b/docs/tools/qemu-trace-stap.rst
@@ -46,19 +46,19 @@ The following commands are valid:
any of the listed names. If no *PATTERN* is given, the all possible
probes will be listed.
- For example, to list all probes available in the ``qemu-system-x86_64``
+ For example, to list all probes available in the ``qemu-kvm``
binary:
::
- $ qemu-trace-stap list qemu-system-x86_64
+ $ qemu-trace-stap list qemu-kvm
To filter the list to only cover probes related to QEMU's cryptographic
subsystem, in a binary outside ``$PATH``
::
- $ qemu-trace-stap list /opt/qemu/4.0.0/bin/qemu-system-x86_64 'qcrypto*'
+ $ qemu-trace-stap list /opt/qemu/4.0.0/bin/qemu-kvm 'qcrypto*'
.. option:: run OPTIONS BINARY PATTERN...
@@ -90,18 +90,18 @@ The following commands are valid:
Restrict the tracing session so that it only triggers for the process
identified by *PID*.
- For example, to monitor all processes executing ``qemu-system-x86_64``
+ For example, to monitor all processes executing ``qemu-kvm``
as found on ``$PATH``, displaying all I/O related probes:
::
- $ qemu-trace-stap run qemu-system-x86_64 'qio*'
+ $ qemu-trace-stap run qemu-kvm 'qio*'
To monitor only the QEMU process with PID 1732
::
- $ qemu-trace-stap run --pid=1732 qemu-system-x86_64 'qio*'
+ $ qemu-trace-stap run --pid=1732 qemu-kvm 'qio*'
To monitor QEMU processes running an alternative binary outside of
``$PATH``, displaying verbose information about setup of the
@@ -109,7 +109,7 @@ The following commands are valid:
::
- $ qemu-trace-stap -v run /opt/qemu/4.0.0/qemu-system-x86_64 'qio*'
+ $ qemu-trace-stap -v run /opt/qemu/4.0.0/qemu-kvm 'qio*'
See also
--------
diff --git a/qemu-options.hx b/qemu-options.hx
index ae2c6dbbfc..94c4a8dbaf 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -3150,11 +3150,11 @@ SRST
::
- qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
- -numa node,memdev=mem \
- -chardev socket,id=chr0,path=/path/to/socket \
- -netdev type=vhost-user,id=net0,chardev=chr0 \
- -device virtio-net-pci,netdev=net0
+ qemu-kvm -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
+ -numa node,memdev=mem \
+ -chardev socket,id=chr0,path=/path/to/socket \
+ -netdev type=vhost-user,id=net0,chardev=chr0 \
+ -device virtio-net-pci,netdev=net0
``-netdev vhost-vdpa,vhostdev=/path/to/dev``
Establish a vhost-vdpa netdev.
--
2.27.0

66
SOURCES/0017-virtio-scsi-Reject-scsi-cd-if-data-plane-enabled-RHE.patch
Unescape View File

@ -0,0 +1,66 @@
From d95768c039a2bf6b68422f83a8d55dad41bd3181 Mon Sep 17 00:00:00 2001
From: Fam Zheng <famz@redhat.com>
Date: Wed, 14 Jun 2017 15:37:01 +0200
Subject: virtio-scsi: Reject scsi-cd if data plane enabled [RHEL only]
RH-Author: Fam Zheng <famz@redhat.com>
Message-id: <20170614153701.14757-1-famz@redhat.com>
Patchwork-id: 75613
O-Subject: [RHV-7.4 qemu-kvm-rhev PATCH v3] virtio-scsi: Reject scsi-cd if data plane enabled [RHEL only]
Bugzilla: 1378816
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Max Reitz <mreitz@redhat.com>
We need a fix for RHEL 7.4 and 7.3.z, but unfortunately upstream isn't
ready. If it were, the changes will be too invasive. To have an idea:
https://lists.gnu.org/archive/html/qemu-devel/2017-05/msg05400.html
is an incomplete attempt to fix part of the issue, and the remaining
work unfortunately involve even more complex changes.
As a band-aid, this partially reverts the effect of ef8875b
(virtio-scsi: Remove op blocker for dataplane, since v2.7). We cannot
simply revert that commit as a whole because we already shipped it in
qemu-kvm-rhev 7.3, since when, block jobs has been possible. We should
only block what has been broken. Also, faithfully reverting the above
commit means adding back the removed op blocker, but that is not enough,
because it still crashes when inserting media into an initially empty
scsi-cd.
All in all, scsi-cd on virtio-scsi-dataplane has basically been unusable
unless the scsi-cd never enters an empty state, so, disable it
altogether. Otherwise it would be much more difficult to avoid
crashing.
Signed-off-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
Signed-off-by: Danilo C. L. de Paula <ddepaula@redhat.com>
---
hw/scsi/virtio-scsi.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
index 51fd09522a..a35257c35a 100644
--- a/hw/scsi/virtio-scsi.c
+++ b/hw/scsi/virtio-scsi.c
@@ -896,6 +896,15 @@ static void virtio_scsi_hotplug(HotplugHandler *hotplug_dev, DeviceState *dev,
AioContext *old_context;
int ret;
+ /* XXX: Remove this check once block backend is capable of handling
+ * AioContext change upon eject/insert.
+ * s->ctx is NULL if ioeventfd is off, s->ctx is qemu_get_aio_context() if
+ * data plane is not used, both cases are safe for scsi-cd. */
+ if (s->ctx && s->ctx != qemu_get_aio_context() &&
+ object_dynamic_cast(OBJECT(dev), "scsi-cd")) {
+ error_setg(errp, "scsi-cd is not supported by data plane");
+ return;
+ }
if (s->ctx && !s->dataplane_fenced) {
if (blk_op_is_blocked(sd->conf.blk, BLOCK_OP_TYPE_DATAPLANE, errp)) {
return;
--
2.27.0

60
SOURCES/0018-BZ1653590-Require-at-least-64kiB-pages-for-downstrea.patch
Unescape View File

@ -0,0 +1,60 @@
From 92bb62c47eab021f8dabecd09b5fbc1706e6a29c Mon Sep 17 00:00:00 2001
From: David Gibson <dgibson@redhat.com>
Date: Wed, 6 Feb 2019 03:58:56 +0000
Subject: BZ1653590: Require at least 64kiB pages for downstream guests & hosts
RH-Author: David Gibson <dgibson@redhat.com>
Message-id: <20190206035856.19058-1-dgibson@redhat.com>
Patchwork-id: 84246
O-Subject: [RHELAV-8.0/rhel qemu-kvm PATCH] BZ1653590: Require at least 64kiB pages for downstream guests & hosts
Bugzilla: 1653590
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
RH-Acked-by: Serhii Popovych <spopovyc@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
Most current POWER guests require 64kiB page support, so that's the default
for the cap-hpt-max-pagesize option in qemu which limits available guest
page sizes. We warn if the value is set smaller than that, but don't
outright fail upstream, because we need to allow for the possibility of
guest (and/or host) kernels configured for 4kiB page sizes.
Downstream, however, we simply don't support 4kiB pagesize configured
kernels in guest or host, so we can have qemu simply error out in this
situation.
Testing: Attempted to start a guest with cap-hpt-max-page-size=4k and verified
it failed immediately with a qemu error
Signed-off-by: David Gibson <dgibson@redhat.com>
Signed-off-by: Danilo C. L. de Paula <ddepaula@redhat.com>
---
hw/ppc/spapr_caps.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/hw/ppc/spapr_caps.c b/hw/ppc/spapr_caps.c
index ed7c077a0d..48a8efe678 100644
--- a/hw/ppc/spapr_caps.c
+++ b/hw/ppc/spapr_caps.c
@@ -332,12 +332,19 @@ bool spapr_check_pagesize(SpaprMachineState *spapr, hwaddr pagesize,
static void cap_hpt_maxpagesize_apply(SpaprMachineState *spapr,
uint8_t val, Error **errp)
{
+#if 0 /* disabled for RHEL */
if (val < 12) {
error_setg(errp, "Require at least 4kiB hpt-max-page-size");
return;
} else if (val < 16) {
warn_report("Many guests require at least 64kiB hpt-max-page-size");
}
+#else /* Only page sizes >=64kiB supported for RHEL */
+ if (val < 16) {
+ error_setg(errp, "Require at least 64kiB hpt-max-page-size");
+ return;
+ }
+#endif
spapr_check_pagesize(spapr, qemu_minrampagesize(), errp);
}
--
2.27.0

53
SOURCES/0019-compat-Update-hw_compat_rhel_8_5.patch
Unescape View File

@ -0,0 +1,53 @@
From a9b5da617c29f48199cbea08d6a1c083877dce10 Mon Sep 17 00:00:00 2001
From: Laurent Vivier <lvivier@redhat.com>
Date: Mon, 15 Nov 2021 14:22:29 +0100
Subject: compat: Update hw_compat_rhel_8_5
RH-Author: Laurent Vivier <lvivier@redhat.com>
RH-MergeRequest: 66: redhat: Update pseries-rhel8.5.0 machine type
RH-Commit: [1/2] 232f2ad2b29d250fbdb8fcea9d814704c575ba2b
RH-Bugzilla: 2022608
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Greg Kurz <gkurz@redhat.com>
Add properties from hw_compat_6_1 as it already includes the ones from
hw_compat_6_0. Add a lately added property from 6.0 too.
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
--
Rebase notes (6.2.0 rc3):
- Included compatc changes introduced in RC2
---
hw/core/machine.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/hw/core/machine.c b/hw/core/machine.c
index 62febde5aa..736c765c30 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -38,7 +38,7 @@
#include "hw/virtio/virtio-pci.h"
/*
- * Mostly the same as hw_compat_6_0
+ * Mostly the same as hw_compat_6_0 and hw_compat_6_1
*/
GlobalProperty hw_compat_rhel_8_5[] = {
/* hw_compat_rhel_8_5 from hw_compat_6_0 */
@@ -51,6 +51,12 @@ GlobalProperty hw_compat_rhel_8_5[] = {
{ "e1000", "init-vet", "off" },
/* hw_compat_rhel_8_5 from hw_compat_6_0 */
{ "e1000e", "init-vet", "off" },
+ /* hw_compat_rhel_8_5 from hw_compat_6_0 */
+ { "vhost-vsock-device", "seqpacket", "off" },
+ /* hw_compat_rhel_8_5 from hw_compat_6_1 */
+ { "vhost-user-vsock-device", "seqpacket", "off" },
+ /* hw_compat_rhel_8_5 from hw_compat_6_1 */
+ { "nvme-ns", "shared", "off" },
};
const size_t hw_compat_rhel_8_5_len = G_N_ELEMENTS(hw_compat_rhel_8_5);
--
2.27.0

43
SOURCES/0020-redhat-Update-pseries-rhel8.5.0-machine-type.patch
Unescape View File

@ -0,0 +1,43 @@
From 82358c35f04f026820b3907069a6c19cd95b654d Mon Sep 17 00:00:00 2001
From: Laurent Vivier <lvivier@redhat.com>
Date: Mon, 15 Nov 2021 14:25:33 +0100
Subject: redhat: Update pseries-rhel8.5.0 machine type
RH-Author: Laurent Vivier <lvivier@redhat.com>
RH-MergeRequest: 66: redhat: Update pseries-rhel8.5.0 machine type
RH-Commit: [2/2] 36f7ad1ea56baaaecb139875ad0a90a6470196be
RH-Bugzilla: 2022608
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Greg Kurz <gkurz@redhat.com>
`
We don't introduce a new machine type for rhel8.6.0 but we need
to keep compatibility with rhel8.5.0 machine type.
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
---
hw/ppc/spapr.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
index cace86028d..2f27888d8a 100644
--- a/hw/ppc/spapr.c
+++ b/hw/ppc/spapr.c
@@ -5177,10 +5177,14 @@ static void spapr_machine_rhel_default_class_options(MachineClass *mc)
static void spapr_machine_rhel850_class_options(MachineClass *mc)
{
+ SpaprMachineClass *smc = SPAPR_MACHINE_CLASS(mc);
+
/* The default machine type must apply the RHEL specific defaults */
spapr_machine_rhel_default_class_options(mc);
compat_props_add(mc->compat_props, hw_compat_rhel_8_5,
hw_compat_rhel_8_5_len);
+ smc->pre_6_2_numa_affinity = true;
+ mc->smp_props.prefer_sockets = true;
}
DEFINE_SPAPR_MACHINE(rhel850, "rhel8.5.0", true);
--
2.27.0

51
SOURCES/0021-redhat-virt-rhel8.5.0-Update-machine-type-compatibil.patch
Unescape View File

@ -0,0 +1,51 @@
From ce73e939b993cc6be170cdb5d3f2068270593f2b Mon Sep 17 00:00:00 2001
From: Eric Auger <eric.auger@redhat.com>
Date: Tue, 16 Nov 2021 17:03:07 +0100
Subject: redhat: virt-rhel8.5.0: Update machine type compatibility for QEMU
6.2.0 update
RH-Author: Eric Auger <eric.auger@redhat.com>
RH-MergeRequest: 75: redhat: virt-rhel8.5.0: Update machine type compatibility for QEMU 6.2.0 update
RH-Commit: [21/21] f027d13654944e3d34e3356affe7af952eec2bed
RH-Bugzilla: 2022607
RH-Acked-by: Gavin Shan <gshan@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Andrew Jones <drjones@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
To keep compatibility with 8.5-AV machine type we need to
turn few new options on by default:
smp_props.prefer_sockets, no_cpu_topology, no_tcg_its
TESTED: migrate from rhel-av-8.5.0 to rhel-8.6.0 and vice-versa
with upstream fix: 33a0c404fb hw/intc/arm_gicv3_its: Revert version
increments in vmstate_its
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
hw/arm/virt.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index c77d26ab13..e8941afd01 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -3225,8 +3225,13 @@ type_init(rhel_machine_init);
static void rhel850_virt_options(MachineClass *mc)
{
+ VirtMachineClass *vmc = VIRT_MACHINE_CLASS(OBJECT_CLASS(mc));
+
compat_props_add(mc->compat_props, arm_rhel_compat, arm_rhel_compat_len);
compat_props_add(mc->compat_props, hw_compat_rhel_8_5, hw_compat_rhel_8_5_len);
+ mc->smp_props.prefer_sockets = true;
+ vmc->no_cpu_topology = true;
+ vmc->no_tcg_its = true;
}
DEFINE_RHEL_MACHINE_AS_LATEST(8, 5, 0)
--
2.27.0

45
SOURCES/0022-Fix-virtio-net-pci-vectors-compat.patch
Unescape View File

@ -0,0 +1,45 @@
From f9643b6934657292aae0b830627b1e5f9b8cbaa1 Mon Sep 17 00:00:00 2001
From: Eduardo Habkost <ehabkost@redhat.com>
Date: Tue, 19 Oct 2021 13:17:06 -0400
Subject: Fix virtio-net-pci* "vectors" compat
RH-Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-MergeRequest: 77: 8.6/6.2 mt fixes
RH-Commit: [21/23] 8ad581932275d2698a99f31bec40b14f1dbd3d2e
RH-Bugzilla: 2026443
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
hw_compat_rhel_8_4 has an issue: it affects only "virtio-net-pci"
but not "virtio-net-pci-transitional" and
"virtio-net-pci-non-transitional". The solution is to use the
"virtio-net-pci-base" type in compat_props.
An equivalent fix will be submitted for hw_compat_5_2 upstream.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
(cherry picked from commit d45823ab0d0138b2fbaf2ed1e1896d2052f3ccb3)
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
hw/core/machine.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/hw/core/machine.c b/hw/core/machine.c
index 736c765c30..024b025fc2 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -71,7 +71,11 @@ GlobalProperty hw_compat_rhel_8_4[] = {
/* hw_compat_rhel_8_4 from hw_compat_5_2 */
{ "virtio-blk-device", "report-discard-granularity", "off" },
/* hw_compat_rhel_8_4 from hw_compat_5_2 */
- { "virtio-net-pci", "vectors", "3"},
+ /*
+ * Upstream incorrectly had "virtio-net-pci" instead of "virtio-net-pci-base",
+ * (https://bugzilla.redhat.com/show_bug.cgi?id=1999141)
+ */
+ { "virtio-net-pci-base", "vectors", "3"},
};
const size_t hw_compat_rhel_8_4_len = G_N_ELEMENTS(hw_compat_rhel_8_4);
--
2.27.0

73
SOURCES/0023-x86-rhel-machine-types-Add-pc_rhel_8_5_compat.patch
Unescape View File

@ -0,0 +1,73 @@
From 7ad8814e583dcc7dc23e3e8398570243b8f176a1 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Date: Tue, 23 Nov 2021 17:57:42 +0000
Subject: x86/rhel machine types: Add pc_rhel_8_5_compat
RH-Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-MergeRequest: 77: 8.6/6.2 mt fixes
RH-Commit: [22/23] 8bf555c5d78f344b97ffd5c888c7a7bed592d9d0
RH-Bugzilla: 2026443
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
Add pc_rhel_8_5_compat as the merge of pc_compat_6_1 and pc_compat_6_0
(since 8.5 was based on 6.0).
Note, x-keep-pci-slot-hpc flipped back and forward, leaving it out
looks like it leaves us with the original.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
hw/i386/pc.c | 21 +++++++++++++++++++++
include/hw/i386/pc.h | 3 +++
2 files changed, 24 insertions(+)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index e8109954ca..4c08a1971c 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -387,6 +387,27 @@ GlobalProperty pc_rhel_compat[] = {
};
const size_t pc_rhel_compat_len = G_N_ELEMENTS(pc_rhel_compat);
+GlobalProperty pc_rhel_8_5_compat[] = {
+ /* pc_rhel_8_5_compat from pc_compat_6_0 */
+ { "qemu64" "-" TYPE_X86_CPU, "family", "6" },
+ /* pc_rhel_8_5_compat from pc_compat_6_0 */
+ { "qemu64" "-" TYPE_X86_CPU, "model", "6" },
+ /* pc_rhel_8_5_compat from pc_compat_6_0 */
+ { "qemu64" "-" TYPE_X86_CPU, "stepping", "3" },
+ /* pc_rhel_8_5_compat from pc_compat_6_0 */
+ { TYPE_X86_CPU, "x-vendor-cpuid-only", "off" },
+ /* pc_rhel_8_5_compat from pc_compat_6_0 */
+ { "ICH9-LPC", ACPI_PM_PROP_ACPI_PCIHP_BRIDGE, "off" },
+
+ /* pc_rhel_8_5_compat from pc_compat_6_1 */
+ { TYPE_X86_CPU, "hv-version-id-build", "0x1bbc" },
+ /* pc_rhel_8_5_compat from pc_compat_6_1 */
+ { TYPE_X86_CPU, "hv-version-id-major", "0x0006" },
+ /* pc_rhel_8_5_compat from pc_compat_6_1 */
+ { TYPE_X86_CPU, "hv-version-id-minor", "0x0001" },
+};
+const size_t pc_rhel_8_5_compat_len = G_N_ELEMENTS(pc_rhel_8_5_compat);
+
GlobalProperty pc_rhel_8_4_compat[] = {
/* pc_rhel_8_4_compat from pc_compat_5_2 */
{ "ICH9-LPC", "x-smi-cpu-hotunplug", "off" },
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index d0544ee119..9e8bfb69f8 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -286,6 +286,9 @@ extern const size_t pc_compat_1_4_len;
extern GlobalProperty pc_rhel_compat[];
extern const size_t pc_rhel_compat_len;
+extern GlobalProperty pc_rhel_8_5_compat[];
+extern const size_t pc_rhel_8_5_compat_len;
+
extern GlobalProperty pc_rhel_8_4_compat[];
extern const size_t pc_rhel_8_4_compat_len;
--
2.27.0

54
SOURCES/0024-x86-rhel-machine-types-Wire-compat-into-q35-and-i440.patch
Unescape View File

@ -0,0 +1,54 @@
From 7bd99eebadfdbea6a76585b526e7cab1ee8b1fde Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Date: Tue, 23 Nov 2021 18:07:49 +0000
Subject: x86/rhel machine types: Wire compat into q35 and i440fx
RH-Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-MergeRequest: 77: 8.6/6.2 mt fixes
RH-Commit: [23/23] fc3861aeccc943b434231193ef45ffbc0b3cf6c6
RH-Bugzilla: 2026443
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
Wire the pc_rhel_8_5 compat data into both piix and q35
to keep the existing machine types compatible.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Miroslav Rezanina <mrezanin@redhat.com>
---
hw/i386/pc_piix.c | 4 ++++
hw/i386/pc_q35.c | 4 ++++
2 files changed, 8 insertions(+)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 2885edffe9..37fab00733 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -1040,6 +1040,10 @@ static void pc_machine_rhel760_options(MachineClass *m)
pcmc->kvmclock_create_always = false;
/* From pc_i440fx_5_1_machine_options() */
pcmc->pci_root_uid = 1;
+ compat_props_add(m->compat_props, hw_compat_rhel_8_5,
+ hw_compat_rhel_8_5_len);
+ compat_props_add(m->compat_props, pc_rhel_8_5_compat,
+ pc_rhel_8_5_compat_len);
compat_props_add(m->compat_props, hw_compat_rhel_8_4,
hw_compat_rhel_8_4_len);
compat_props_add(m->compat_props, pc_rhel_8_4_compat,
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index c67418b6a9..78876e1101 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -658,6 +658,10 @@ static void pc_q35_machine_rhel850_options(MachineClass *m)
m->desc = "RHEL-8.5.0 PC (Q35 + ICH9, 2009)";
pcmc->smbios_stream_product = "RHEL-AV";
pcmc->smbios_stream_version = "8.5.0";
+ compat_props_add(m->compat_props, hw_compat_rhel_8_5,
+ hw_compat_rhel_8_5_len);
+ compat_props_add(m->compat_props, pc_rhel_8_5_compat,
+ pc_rhel_8_5_compat_len);
}
DEFINE_PC_MACHINE(q35_rhel850, "pc-q35-rhel8.5.0", pc_q35_init_rhel850,
--
2.27.0

58
SOURCES/0025-redhat-Add-s390x-machine-type-compatibility-handling.patch
Unescape View File

@ -0,0 +1,58 @@
From 265a57f2955b7f0b65e3f57f89aa1ff2541d3f73 Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Fri, 26 Nov 2021 09:37:11 +0100
Subject: redhat: Add s390x machine type compatibility handling for the rebase
to v6.2
RH-Author: Thomas Huth <thuth@redhat.com>
RH-MergeRequest: 80: Add s390x machine type compatibility handling for the rebase to v6.2
RH-Commit: [26/26] c45cf594604f6dd23954696b9c84d2025e328d11
RH-Bugzilla: 2022602
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Laurent Vivier <lvivier@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
Add compatibility handling for the rhel8.5.0 machine type (and
recursively older, of course).
Based on the following upstream commits:
463e50da8b - s390x/cpumodel: Bump up QEMU model to a stripped-down IBM z14 GA2
30e398f796 - s390x/cpumodel: Add more feature to gen16 default model
4a0af2930a - machine: Prefer cores over sockets in smp parsing since 6.2
2b52619994 - machine: Move smp_prefer_sockets to struct SMPCompatProps
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
hw/s390x/s390-virtio-ccw.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index 181856e6cf..cf13c457d6 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -1105,11 +1105,21 @@ DEFINE_CCW_MACHINE(2_4, "2.4", false);
static void ccw_machine_rhel850_instance_options(MachineState *machine)
{
+ static const S390FeatInit qemu_cpu_feat = { S390_FEAT_LIST_QEMU_V6_0 };
+
+ s390_set_qemu_cpu_model(0x2964, 13, 2, qemu_cpu_feat);
+
+ s390_cpudef_featoff_greater(16, 1, S390_FEAT_NNPA);
+ s390_cpudef_featoff_greater(16, 1, S390_FEAT_VECTOR_PACKED_DECIMAL_ENH2);
+ s390_cpudef_featoff_greater(16, 1, S390_FEAT_BEAR_ENH);
+ s390_cpudef_featoff_greater(16, 1, S390_FEAT_RDP);
+ s390_cpudef_featoff_greater(16, 1, S390_FEAT_PAI);
}
static void ccw_machine_rhel850_class_options(MachineClass *mc)
{
compat_props_add(mc->compat_props, hw_compat_rhel_8_5, hw_compat_rhel_8_5_len);
+ mc->smp_props.prefer_sockets = true;
}
DEFINE_CCW_MACHINE(rhel850, "rhel8.5.0", true);
--
2.27.0

1
SOURCES/81-kvm-rhel.rules
Unescape View File

@ -0,0 +1 @@
DEVPATH=="*/kvm", ACTION=="change", RUN+="/lib/udev/udev-kvm-check $env{COUNT} $env{EVENT}"

5
SOURCES/85-kvm.preset
Unescape View File

@ -0,0 +1,5 @@
# Enable kvm-setup by default. This can have odd side effects on
# PowerNV systems that aren't intended as KVM hosts, but at present we
# only support RHEL on PowerNV for the purpose of being a RHEV host.
enable kvm-setup.service

10
SOURCES/95-kvm-memlock.conf
Unescape View File

@ -0,0 +1,10 @@
# The KVM HV implementation on Power can require a significant amount
# of unswappable memory (about half of which also needs to be host
# physically contiguous) to hold the guest's Hash Page Table (HPT) -
# roughly 1/64th of the guest's RAM size, minimum 16MiB.
#
# These limits allow unprivileged users to start smallish VMs, such as
# those used by libguestfs.
#
* hard memlock 65536
* soft memlock 65536

2
SOURCES/99-qemu-guest-agent.rules
Unescape View File

@ -0,0 +1,2 @@
SUBSYSTEM=="virtio-ports", ATTR{name}=="org.qemu.guest_agent.0", \
TAG+="systemd" ENV{SYSTEMD_WANTS}="qemu-guest-agent.service"

39
SOURCES/README.tests
Unescape View File

@ -0,0 +1,39 @@
qemu-kvm-tests README
=====================
The qemu-kvm-tests rpm contains tests that can be used to verify the
functionality of the installed qemu-kvm package
When installed, the files from this rpm will be arranged in the following
directory structure
tests-src/
├── README
├── scripts
│   ├── qemu.py
│   └── qmp
└── tests
├── acceptance
├── Makefile.include
└── qemu-iotests
The tests/ directory within the tests-src/ directory is setup to remain a copy
of a subset of the tests/ directory from the QEMU source tree
The avocado_qemu tests and qemu-iotests, along with files required for the
execution of the avocado_qemu tests (scripts/qemu.py and scripts/qmp/) will be
installed in a new location - /usr/lib64/qemu-kvm/tests-src/
avocado_qemu tests:
The avocado_qemu tests can be executed by running the following avocado command:
avocado run -p qemu_bin=/usr/libexec/qemu-kvm /usr/lib64/qemu-kvm/tests/acceptance/
Avocado needs to be installed separately using either pip or from source as
Avocado is not being packaged for RHEL-8.
qemu-iotests:
symlinks to corresponding binaries need to be created for QEMU_PROG,
QEMU_IO_PROG, QEMU_IMG_PROG, and QEMU_NBD_PROG before the iotests can be
executed.
The primary purpose of this package is to make these tests available to be
executed as gating tests for the virt module in the RHEL-8 OSCI environment.

1
SOURCES/bridge.conf
Unescape View File

@ -0,0 +1 @@
allow virbr0

13
SOURCES/ksm.service
Unescape View File

@ -0,0 +1,13 @@
[Unit]
Description=Kernel Samepage Merging
ConditionPathExists=/sys/kernel/mm/ksm
[Service]
Type=oneshot
RemainAfterExit=yes
EnvironmentFile=-/etc/sysconfig/ksm
ExecStart=/usr/libexec/ksmctl start
ExecStop=/usr/libexec/ksmctl stop
[Install]
WantedBy=multi-user.target

4
SOURCES/ksm.sysconfig
Unescape View File

@ -0,0 +1,4 @@
# The maximum number of unswappable kernel pages
# which may be allocated by ksm (0 for unlimited)
# If unset, defaults to half of total memory
# KSM_MAX_KERNEL_PAGES=

77
SOURCES/ksmctl.c
Unescape View File

@ -0,0 +1,77 @@
/* Start/stop KSM, for systemd.
* Copyright (C) 2009, 2011 Red Hat, Inc.
* Written by Paolo Bonzini <pbonzini@redhat.com>.
* Based on the original sysvinit script by Dan Kenigsberg <danken@redhat.com>
* This file is distributed under the GNU General Public License, version 2
* or later. */
#include <unistd.h>
#include <stdio.h>
#include <limits.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
#define KSM_MAX_KERNEL_PAGES_FILE "/sys/kernel/mm/ksm/max_kernel_pages"
#define KSM_RUN_FILE "/sys/kernel/mm/ksm/run"
char *program_name;
int usage(void)
{
fprintf(stderr, "Usage: %s {start|stop}\n", program_name);
return 1;
}
int write_value(uint64_t value, char *filename)
{
FILE *fp;
if (!(fp = fopen(filename, "w")) ||
fprintf(fp, "%llu\n", (unsigned long long) value) == EOF ||
fflush(fp) == EOF ||
fclose(fp) == EOF)
return 1;
return 0;
}
uint64_t ksm_max_kernel_pages()
{
char *var = getenv("KSM_MAX_KERNEL_PAGES");
char *endptr;
uint64_t value;
if (var && *var) {
value = strtoll(var, &endptr, 0);
if (value < LLONG_MAX && !*endptr)
return value;
}
/* Unless KSM_MAX_KERNEL_PAGES is set, let KSM munch up to half of
* total memory. */
return sysconf(_SC_PHYS_PAGES) / 2;
}
int start(void)
{
if (access(KSM_MAX_KERNEL_PAGES_FILE, R_OK) >= 0)
write_value(ksm_max_kernel_pages(), KSM_MAX_KERNEL_PAGES_FILE);
return write_value(1, KSM_RUN_FILE);
}
int stop(void)
{
return write_value(0, KSM_RUN_FILE);
}
int main(int argc, char **argv)
{
program_name = argv[0];
if (argc < 2) {
return usage();
} else if (!strcmp(argv[1], "start")) {
return start();
} else if (!strcmp(argv[1], "stop")) {
return stop();
} else {
return usage();
}
}

139
SOURCES/ksmtuned
Unescape View File

@ -0,0 +1,139 @@
#!/bin/bash
#
# Copyright 2009 Red Hat, Inc. and/or its affiliates.
# Released under the GPL
#
# Author: Dan Kenigsberg <danken@redhat.com>
#
# ksmtuned - a simple script that controls whether (and with what vigor) ksm
# should search for duplicated pages.
#
# starts ksm when memory commited to qemu processes exceeds a threshold, and
# make ksm work harder and harder untill memory load falls below that
# threshold.
#
# send SIGUSR1 to this process right after a new qemu process is started, or
# following its death, to retune ksm accordingly
#
# needs testing and ironing. contact danken@redhat.com if something breaks.
if [ -f /etc/ksmtuned.conf ]; then
. /etc/ksmtuned.conf
fi
debug() {
if [ -n "$DEBUG" ]; then
s="`/bin/date`: $*"
[ -n "$LOGFILE" ] && echo "$s" >> "$LOGFILE" || echo "$s"
fi
}
KSM_MONITOR_INTERVAL=${KSM_MONITOR_INTERVAL:-60}
KSM_NPAGES_BOOST=${KSM_NPAGES_BOOST:-300}
KSM_NPAGES_DECAY=${KSM_NPAGES_DECAY:--50}
KSM_NPAGES_MIN=${KSM_NPAGES_MIN:-64}
KSM_NPAGES_MAX=${KSM_NPAGES_MAX:-1250}
# millisecond sleep between ksm scans for 16Gb server. Smaller servers sleep
# more, bigger sleep less.
KSM_SLEEP_MSEC=${KSM_SLEEP_MSEC:-10}
KSM_THRES_COEF=${KSM_THRES_COEF:-20}
KSM_THRES_CONST=${KSM_THRES_CONST:-2048}
total=`awk '/^MemTotal:/ {print $2}' /proc/meminfo`
debug total $total
npages=0
sleep=$[KSM_SLEEP_MSEC * 16 * 1024 * 1024 / total]
[ $sleep -le 10 ] && sleep=10
debug sleep $sleep
thres=$[total * KSM_THRES_COEF / 100]
if [ $KSM_THRES_CONST -gt $thres ]; then
thres=$KSM_THRES_CONST
fi
debug thres $thres
KSMCTL () {
case x$1 in
xstop)
echo 0 > /sys/kernel/mm/ksm/run
;;
xstart)
echo $2 > /sys/kernel/mm/ksm/pages_to_scan
echo $3 > /sys/kernel/mm/ksm/sleep_millisecs
echo 1 > /sys/kernel/mm/ksm/run
;;
esac
}
committed_memory () {
# calculate how much memory is committed to running qemu processes
local pidlist
pidlist=$(pgrep -d ' ' -- '^qemu(-(kvm|system-.+)|:.{1,11})$')
if [ -n "$pidlist" ]; then
ps -p "$pidlist" -o rsz=
fi | awk '{ sum += $1 }; END { print 0+sum }'
}
free_memory () {
awk '/^(MemFree|Buffers|Cached):/ {free += $2}; END {print free}' \
/proc/meminfo
}
increase_npages() {
local delta
delta=${1:-0}
npages=$[npages + delta]
if [ $npages -lt $KSM_NPAGES_MIN ]; then
npages=$KSM_NPAGES_MIN
elif [ $npages -gt $KSM_NPAGES_MAX ]; then
npages=$KSM_NPAGES_MAX
fi
echo $npages
}
adjust () {
local free committed
free=`free_memory`
committed=`committed_memory`
debug committed $committed free $free
if [ $[committed + thres] -lt $total -a $free -gt $thres ]; then
KSMCTL stop
debug "$[committed + thres] < $total and free > $thres, stop ksm"
return 1
fi
debug "$[committed + thres] > $total, start ksm"
if [ $free -lt $thres ]; then
npages=`increase_npages $KSM_NPAGES_BOOST`
debug "$free < $thres, boost"
else
npages=`increase_npages $KSM_NPAGES_DECAY`
debug "$free > $thres, decay"
fi
KSMCTL start $npages $sleep
debug "KSMCTL start $npages $sleep"
return 0
}
function nothing () {
:
}
loop () {
trap nothing SIGUSR1
while true
do
sleep $KSM_MONITOR_INTERVAL &
wait $!
adjust
done
}
PIDFILE=${PIDFILE-/var/run/ksmtune.pid}
if touch "$PIDFILE"; then
loop &
echo $! > "$PIDFILE"
fi

21
SOURCES/ksmtuned.conf
Unescape View File

@ -0,0 +1,21 @@
# Configuration file for ksmtuned.
# How long ksmtuned should sleep between tuning adjustments
# KSM_MONITOR_INTERVAL=60
# Millisecond sleep between ksm scans for 16Gb server.
# Smaller servers sleep more, bigger sleep less.
# KSM_SLEEP_MSEC=10
# KSM_NPAGES_BOOST=300
# KSM_NPAGES_DECAY=-50
# KSM_NPAGES_MIN=64
# KSM_NPAGES_MAX=1250
# KSM_THRES_COEF=20
# KSM_THRES_CONST=2048
# uncomment the following if you want ksmtuned debug info
# LOGFILE=/var/log/ksmtuned
# DEBUG=1

12
SOURCES/ksmtuned.service
Unescape View File

@ -0,0 +1,12 @@
[Unit]
Description=Kernel Samepage Merging (KSM) Tuning Daemon
After=ksm.service
Requires=ksm.service
[Service]
ExecStart=/usr/sbin/ksmtuned
ExecReload=/bin/kill -USR1 $MAINPID
Type=forking
[Install]
WantedBy=multi-user.target

87
SOURCES/kvm-Add-dirty-sync-missed-zero-copy-migration-stat.patch
Unescape View File

@ -0,0 +1,87 @@
From cd49a32e9c9e33efc51652b68180a07683814b4d Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 11 Jul 2022 18:11:12 -0300
Subject: [PATCH 4/9] Add dirty-sync-missed-zero-copy migration stat
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 201: Zero-copy-send fixes + improvements
RH-Commit: [4/8] 56cce61cf95aafc8dafae7531b43c166084abfec
RH-Bugzilla: 2110203
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220711211112.18951-3-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit cf20c897338067ab4b70a4596fdccaf90c7e29a1)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
migration/migration.c | 2 ++
monitor/hmp-cmds.c | 5 +++++
qapi/migration.json | 7 ++++++-
3 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/migration/migration.c b/migration/migration.c
index e100b30f00..952a26c5c2 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -1012,6 +1012,8 @@ static void populate_ram_info(MigrationInfo *info, MigrationState *s)
info->ram->normal_bytes = ram_counters.normal * page_size;
info->ram->mbps = s->mbps;
info->ram->dirty_sync_count = ram_counters.dirty_sync_count;
+ info->ram->dirty_sync_missed_zero_copy =
+ ram_counters.dirty_sync_missed_zero_copy;
info->ram->postcopy_requests = ram_counters.postcopy_requests;
info->ram->page_size = page_size;
info->ram->multifd_bytes = ram_counters.multifd_bytes;
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
index 8c384dc1b2..f7216ab5d0 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
@@ -305,6 +305,11 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict)
monitor_printf(mon, "postcopy ram: %" PRIu64 " kbytes\n",
info->ram->postcopy_bytes >> 10);
}
+ if (info->ram->dirty_sync_missed_zero_copy) {
+ monitor_printf(mon,
+ "Zero-copy-send fallbacks happened: %" PRIu64 " times\n",
+ info->ram->dirty_sync_missed_zero_copy);
+ }
}
if (info->has_disk) {
diff --git a/qapi/migration.json b/qapi/migration.json
index c8ec260ab0..94bc5c69db 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -55,6 +55,10 @@
# @postcopy-bytes: The number of bytes sent during the post-copy phase
# (since 7.0).
#
+# @dirty-sync-missed-zero-copy: Number of times dirty RAM synchronization could
+# not avoid copying dirty pages. This is between
+# 0 and @dirty-sync-count * @multifd-channels.
+# (since 7.1)
# Since: 0.14
##
{ 'struct': 'MigrationStats',
@@ -65,7 +69,8 @@
'postcopy-requests' : 'int', 'page-size' : 'int',
'multifd-bytes' : 'uint64', 'pages-per-second' : 'uint64',
'precopy-bytes' : 'uint64', 'downtime-bytes' : 'uint64',
- 'postcopy-bytes' : 'uint64' } }
+ 'postcopy-bytes' : 'uint64',
+ 'dirty-sync-missed-zero-copy' : 'uint64' } }
##
# @XBZRLECacheStats:
--
2.31.1

28
SOURCES/kvm-Enable-SGX-RH-Only.patch
Unescape View File

@ -0,0 +1,28 @@
From db6e042fe4fdc1a1bbf562a46b15d4d8e33e2fa6 Mon Sep 17 00:00:00 2001
From: Paul Lai <plai@redhat.com>
Date: Tue, 25 Jan 2022 15:16:22 -0500
Subject: [PATCH 4/7] Enable SGX -- RH Only
RH-Author: Paul Lai <None>
RH-MergeRequest: 111: numa: Enable numa for SGX EPC sections
RH-Commit: [4/5] cea874f29984897ef1232fb7749c13203c888034
RH-Bugzilla: 1518984
RH-Acked-by: Paolo Bonzini <None>
RH-Acked-by: Bandan Das <None>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
---
configs/devices/x86_64-softmmu/x86_64-rh-devices.mak | 1 +
1 file changed, 1 insertion(+)
diff --git a/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak b/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
index ddf036f042..fdbbdf9742 100644
--- a/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
+++ b/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
@@ -102,3 +102,4 @@ CONFIG_TPM_CRB=y
CONFIG_TPM_TIS_ISA=y
CONFIG_TPM_EMULATOR=y
CONFIG_TPM_PASSTHROUGH=y
+CONFIG_SGX=y
--
2.27.0

82
SOURCES/kvm-KVM-keep-track-of-running-ioctls.patch
Unescape View File

@ -0,0 +1,82 @@
From 9bacf8c4104ff3cff2e0e2c2179ec4fda633167f Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Mon, 16 Jan 2023 07:51:08 -0500
Subject: [PATCH 05/11] KVM: keep track of running ioctls
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 247: accel: introduce accelerator blocker API
RH-Bugzilla: 2161188
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Vitaly Kuznetsov <vkuznets@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [2/3] 357508389e2a0fd996206b406e9e235e50b5f0b6
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2161188
commit a27dd2de68f37ba96fe164a42121daa5f0750afc
Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Fri Nov 11 10:47:57 2022 -0500
KVM: keep track of running ioctls
Using the new accel-blocker API, mark where ioctls are being called
in KVM. Next, we will implement the critical section that will take
care of performing memslots modifications atomically, therefore
preventing any new ioctl from running and allowing the running ones
to finish.
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Message-Id: <20221111154758.1372674-3-eesposit@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
accel/kvm/kvm-all.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 8f2a53438f..221aadfda7 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2337,6 +2337,7 @@ static int kvm_init(MachineState *ms)
assert(TARGET_PAGE_SIZE <= qemu_real_host_page_size);
s->sigmask_len = 8;
+ accel_blocker_init();
#ifdef KVM_CAP_SET_GUEST_DEBUG
QTAILQ_INIT(&s->kvm_sw_breakpoints);
@@ -3018,7 +3019,9 @@ int kvm_vm_ioctl(KVMState *s, int type, ...)
va_end(ap);
trace_kvm_vm_ioctl(type, arg);
+ accel_ioctl_begin();
ret = ioctl(s->vmfd, type, arg);
+ accel_ioctl_end();
if (ret == -1) {
ret = -errno;
}
@@ -3036,7 +3039,9 @@ int kvm_vcpu_ioctl(CPUState *cpu, int type, ...)
va_end(ap);
trace_kvm_vcpu_ioctl(cpu->cpu_index, type, arg);
+ accel_cpu_ioctl_begin(cpu);
ret = ioctl(cpu->kvm_fd, type, arg);
+ accel_cpu_ioctl_end(cpu);
if (ret == -1) {
ret = -errno;
}
@@ -3054,7 +3059,9 @@ int kvm_device_ioctl(int fd, int type, ...)
va_end(ap);
trace_kvm_device_ioctl(fd, type, arg);
+ accel_ioctl_begin();
ret = ioctl(fd, type, arg);
+ accel_ioctl_end();
if (ret == -1) {
ret = -errno;
}
--
2.37.3

109
SOURCES/kvm-KVM-x86-workaround-invalid-CPUID-0xD-9-info-on-some-.patch
Unescape View File

@ -0,0 +1,109 @@
From ea5299b5dde7d0b6b2f93cb646e6a24c9f105466 Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 23 Mar 2022 12:33:25 +0100
Subject: [PATCH 13/24] KVM: x86: workaround invalid CPUID[0xD,9] info on some
AMD processors
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Paul Lai <plai@redhat.com>
RH-MergeRequest: 176: Enable KVM AMX support
RH-Commit: [13/13] 38f147c911258e84e01336271ebd23a1c24371fc
RH-Bugzilla: 1916415
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Some AMD processors expose the PKRU extended save state even if they do not have
the related PKU feature in CPUID. Worse, when they do they report a size of
64, whereas the expected size of the PKRU extended save state is 8, therefore
the esa->size == eax assertion does not hold.
The state is already ignored by KVM_GET_SUPPORTED_CPUID because it
was not enabled in the host XCR0. However, QEMU kvm_cpu_xsave_init()
runs before QEMU invokes arch_prctl() to enable dynamically-enabled
save states such as XTILEDATA, and KVM_GET_SUPPORTED_CPUID hides save
states that have yet to be enabled. Therefore, kvm_cpu_xsave_init()
needs to consult the host CPUID instead of KVM_GET_SUPPORTED_CPUID,
and dies with an assertion failure.
When setting up the ExtSaveArea array to match the host, ignore features that
KVM does not report as supported. This will cause QEMU to skip the incorrect
CPUID leaf instead of tripping the assertion.
Closes: https://gitlab.com/qemu-project/qemu/-/issues/916
Reported-by: Daniel P. Berrangé <berrange@redhat.com>
Analyzed-by: Yang Zhong <yang.zhong@intel.com>
Reported-by: Peter Krempa <pkrempa@redhat.com>
Tested-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 58f7db26f21c690cf9a669c314cfd7371506084a)
Signed-off-by: Paul Lai <plai@redhat.com>
---
target/i386/cpu.c | 4 ++--
target/i386/cpu.h | 2 ++
target/i386/kvm/kvm-cpu.c | 19 ++++++++++++-------
3 files changed, 16 insertions(+), 9 deletions(-)
diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 09e08f7f38..0543b846ff 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -4980,8 +4980,8 @@ CpuDefinitionInfoList *qmp_query_cpu_definitions(Error **errp)
return cpu_list;
}
-static uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
- bool migratable_only)
+uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
+ bool migratable_only)
{
FeatureWordInfo *wi = &feature_word_info[w];
uint64_t r = 0;
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 8ab2a4042a..006b735fe4 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -604,6 +604,8 @@ typedef enum FeatureWord {
} FeatureWord;
typedef uint64_t FeatureWordArray[FEATURE_WORDS];
+uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
+ bool migratable_only);
/* cpuid_features bits */
#define CPUID_FP87 (1U << 0)
diff --git a/target/i386/kvm/kvm-cpu.c b/target/i386/kvm/kvm-cpu.c
index bdc967c484..74c1396a93 100644
--- a/target/i386/kvm/kvm-cpu.c
+++ b/target/i386/kvm/kvm-cpu.c
@@ -99,13 +99,18 @@ static void kvm_cpu_xsave_init(void)
for (i = XSTATE_SSE_BIT + 1; i < XSAVE_STATE_AREA_COUNT; i++) {
ExtSaveArea *esa = &x86_ext_save_areas[i];
- if (esa->size) {
- host_cpuid(0xd, i, &eax, &ebx, &ecx, &edx);
- if (eax != 0) {
- assert(esa->size == eax);
- esa->offset = ebx;
- esa->ecx = ecx;
- }
+ if (!esa->size) {
+ continue;
+ }
+ if ((x86_cpu_get_supported_feature_word(esa->feature, false) & esa->bits)
+ != esa->bits) {
+ continue;
+ }
+ host_cpuid(0xd, i, &eax, &ebx, &ecx, &edx);
+ if (eax != 0) {
+ assert(esa->size == eax);
+ esa->offset = ebx;
+ esa->ecx = ecx;
}
}
}
--
2.35.3

420
SOURCES/kvm-QIOChannel-Add-flags-on-io_writev-and-introduce-io_f.patch
Unescape View File

@ -0,0 +1,420 @@
From 7eeec7c008e947bc3e1fed682791092b408852c6 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Wed, 18 May 2022 02:52:24 -0300
Subject: [PATCH 17/37] QIOChannel: Add flags on io_writev and introduce
io_flush callback
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 191: MSG_ZEROCOPY + Multifd @ rhel8.7
RH-Commit: [17/26] 7bde4e79fd3f76a6cc84d9cacf50420584ddd35c
RH-Bugzilla: 2072049
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Daniel P. Berrangé <berrange@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Add flags to io_writev and introduce io_flush as optional callback to
QIOChannelClass, allowing the implementation of zero copy writes by
subclasses.
How to use them:
- Write data using qio_channel_writev*(...,QIO_CHANNEL_WRITE_FLAG_ZERO_COPY),
- Wait write completion with qio_channel_flush().
Notes:
As some zero copy write implementations work asynchronously, it's
recommended to keep the write buffer untouched until the return of
qio_channel_flush(), to avoid the risk of sending an updated buffer
instead of the buffer state during write.
As io_flush callback is optional, if a subclass does not implement it, then:
- io_flush will return 0 without changing anything.
Also, some functions like qio_channel_writev_full_all() were adapted to
receive a flag parameter. That allows shared code between zero copy and
non-zero copy writev, and also an easier implementation on new flags.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220513062836.965425-3-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit b88651cb4d4fa416fdbb6afaf5b26ec8c035eaad)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
chardev/char-io.c | 2 +-
hw/remote/mpqemu-link.c | 2 +-
include/io/channel.h | 38 +++++++++++++++++++++-
io/channel-buffer.c | 1 +
io/channel-command.c | 1 +
io/channel-file.c | 1 +
io/channel-socket.c | 2 ++
io/channel-tls.c | 1 +
io/channel-websock.c | 1 +
io/channel.c | 49 +++++++++++++++++++++++------
migration/rdma.c | 1 +
scsi/pr-manager-helper.c | 2 +-
tests/unit/test-io-channel-socket.c | 1 +
13 files changed, 88 insertions(+), 14 deletions(-)
diff --git a/chardev/char-io.c b/chardev/char-io.c
index 8ced184160..4451128cba 100644
--- a/chardev/char-io.c
+++ b/chardev/char-io.c
@@ -122,7 +122,7 @@ int io_channel_send_full(QIOChannel *ioc,
ret = qio_channel_writev_full(
ioc, &iov, 1,
- fds, nfds, NULL);
+ fds, nfds, 0, NULL);
if (ret == QIO_CHANNEL_ERR_BLOCK) {
if (offset) {
return offset;
diff --git a/hw/remote/mpqemu-link.c b/hw/remote/mpqemu-link.c
index 7e841820e5..e8f556bd27 100644
--- a/hw/remote/mpqemu-link.c
+++ b/hw/remote/mpqemu-link.c
@@ -69,7 +69,7 @@ bool mpqemu_msg_send(MPQemuMsg *msg, QIOChannel *ioc, Error **errp)
}
if (!qio_channel_writev_full_all(ioc, send, G_N_ELEMENTS(send),
- fds, nfds, errp)) {
+ fds, nfds, 0, errp)) {
ret = true;
} else {
trace_mpqemu_send_io_error(msg->cmd, msg->size, nfds);
diff --git a/include/io/channel.h b/include/io/channel.h
index 88988979f8..c680ee7480 100644
--- a/include/io/channel.h
+++ b/include/io/channel.h
@@ -32,12 +32,15 @@ OBJECT_DECLARE_TYPE(QIOChannel, QIOChannelClass,
#define QIO_CHANNEL_ERR_BLOCK -2
+#define QIO_CHANNEL_WRITE_FLAG_ZERO_COPY 0x1
+
typedef enum QIOChannelFeature QIOChannelFeature;
enum QIOChannelFeature {
QIO_CHANNEL_FEATURE_FD_PASS,
QIO_CHANNEL_FEATURE_SHUTDOWN,
QIO_CHANNEL_FEATURE_LISTEN,
+ QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY,
};
@@ -104,6 +107,7 @@ struct QIOChannelClass {
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp);
ssize_t (*io_readv)(QIOChannel *ioc,
const struct iovec *iov,
@@ -136,6 +140,8 @@ struct QIOChannelClass {
IOHandler *io_read,
IOHandler *io_write,
void *opaque);
+ int (*io_flush)(QIOChannel *ioc,
+ Error **errp);
};
/* General I/O handling functions */
@@ -228,6 +234,7 @@ ssize_t qio_channel_readv_full(QIOChannel *ioc,
* @niov: the length of the @iov array
* @fds: an array of file handles to send
* @nfds: number of file handles in @fds
+ * @flags: write flags (QIO_CHANNEL_WRITE_FLAG_*)
* @errp: pointer to a NULL-initialized error object
*
* Write data to the IO channel, reading it from the
@@ -260,6 +267,7 @@ ssize_t qio_channel_writev_full(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp);
/**
@@ -837,6 +845,7 @@ int qio_channel_readv_full_all(QIOChannel *ioc,
* @niov: the length of the @iov array
* @fds: an array of file handles to send
* @nfds: number of file handles in @fds
+ * @flags: write flags (QIO_CHANNEL_WRITE_FLAG_*)
* @errp: pointer to a NULL-initialized error object
*
*
@@ -846,6 +855,14 @@ int qio_channel_readv_full_all(QIOChannel *ioc,
* to be written, yielding from the current coroutine
* if required.
*
+ * If QIO_CHANNEL_WRITE_FLAG_ZERO_COPY is passed in flags,
+ * instead of waiting for all requested data to be written,
+ * this function will wait until it's all queued for writing.
+ * In this case, if the buffer gets changed between queueing and
+ * sending, the updated buffer will be sent. If this is not a
+ * desired behavior, it's suggested to call qio_channel_flush()
+ * before reusing the buffer.
+ *
* Returns: 0 if all bytes were written, or -1 on error
*/
@@ -853,6 +870,25 @@ int qio_channel_writev_full_all(QIOChannel *ioc,
const struct iovec *iov,
size_t niov,
int *fds, size_t nfds,
- Error **errp);
+ int flags, Error **errp);
+
+/**
+ * qio_channel_flush:
+ * @ioc: the channel object
+ * @errp: pointer to a NULL-initialized error object
+ *
+ * Will block until every packet queued with
+ * qio_channel_writev_full() + QIO_CHANNEL_WRITE_FLAG_ZERO_COPY
+ * is sent, or return in case of any error.
+ *
+ * If not implemented, acts as a no-op, and returns 0.
+ *
+ * Returns -1 if any error is found,
+ * 1 if every send failed to use zero copy.
+ * 0 otherwise.
+ */
+
+int qio_channel_flush(QIOChannel *ioc,
+ Error **errp);
#endif /* QIO_CHANNEL_H */
diff --git a/io/channel-buffer.c b/io/channel-buffer.c
index baa4e2b089..bf52011be2 100644
--- a/io/channel-buffer.c
+++ b/io/channel-buffer.c
@@ -81,6 +81,7 @@ static ssize_t qio_channel_buffer_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelBuffer *bioc = QIO_CHANNEL_BUFFER(ioc);
diff --git a/io/channel-command.c b/io/channel-command.c
index b2a9e27138..5ff1691bad 100644
--- a/io/channel-command.c
+++ b/io/channel-command.c
@@ -258,6 +258,7 @@ static ssize_t qio_channel_command_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelCommand *cioc = QIO_CHANNEL_COMMAND(ioc);
diff --git a/io/channel-file.c b/io/channel-file.c
index c4bf799a80..348a48545e 100644
--- a/io/channel-file.c
+++ b/io/channel-file.c
@@ -114,6 +114,7 @@ static ssize_t qio_channel_file_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelFile *fioc = QIO_CHANNEL_FILE(ioc);
diff --git a/io/channel-socket.c b/io/channel-socket.c
index 606ec97cf7..bfbd64787e 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -525,6 +525,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
@@ -620,6 +621,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
diff --git a/io/channel-tls.c b/io/channel-tls.c
index 2ae1b92fc0..4ce890a538 100644
--- a/io/channel-tls.c
+++ b/io/channel-tls.c
@@ -301,6 +301,7 @@ static ssize_t qio_channel_tls_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelTLS *tioc = QIO_CHANNEL_TLS(ioc);
diff --git a/io/channel-websock.c b/io/channel-websock.c
index 70889bb54d..035dd6075b 100644
--- a/io/channel-websock.c
+++ b/io/channel-websock.c
@@ -1127,6 +1127,7 @@ static ssize_t qio_channel_websock_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelWebsock *wioc = QIO_CHANNEL_WEBSOCK(ioc);
diff --git a/io/channel.c b/io/channel.c
index e8b019dc36..0640941ac5 100644
--- a/io/channel.c
+++ b/io/channel.c
@@ -72,18 +72,32 @@ ssize_t qio_channel_writev_full(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
- if ((fds || nfds) &&
- !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
+ if (fds || nfds) {
+ if (!qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
+ error_setg_errno(errp, EINVAL,
+ "Channel does not support file descriptor passing");
+ return -1;
+ }
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+ error_setg_errno(errp, EINVAL,
+ "Zero Copy does not support file descriptor passing");
+ return -1;
+ }
+ }
+
+ if ((flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) &&
+ !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY)) {
error_setg_errno(errp, EINVAL,
- "Channel does not support file descriptor passing");
+ "Requested Zero Copy feature is not available");
return -1;
}
- return klass->io_writev(ioc, iov, niov, fds, nfds, errp);
+ return klass->io_writev(ioc, iov, niov, fds, nfds, flags, errp);
}
@@ -217,14 +231,14 @@ int qio_channel_writev_all(QIOChannel *ioc,
size_t niov,
Error **errp)
{
- return qio_channel_writev_full_all(ioc, iov, niov, NULL, 0, errp);
+ return qio_channel_writev_full_all(ioc, iov, niov, NULL, 0, 0, errp);
}
int qio_channel_writev_full_all(QIOChannel *ioc,
const struct iovec *iov,
size_t niov,
int *fds, size_t nfds,
- Error **errp)
+ int flags, Error **errp)
{
int ret = -1;
struct iovec *local_iov = g_new(struct iovec, niov);
@@ -237,8 +251,10 @@ int qio_channel_writev_full_all(QIOChannel *ioc,
while (nlocal_iov > 0) {
ssize_t len;
- len = qio_channel_writev_full(ioc, local_iov, nlocal_iov, fds, nfds,
- errp);
+
+ len = qio_channel_writev_full(ioc, local_iov, nlocal_iov, fds,
+ nfds, flags, errp);
+
if (len == QIO_CHANNEL_ERR_BLOCK) {
if (qemu_in_coroutine()) {
qio_channel_yield(ioc, G_IO_OUT);
@@ -277,7 +293,7 @@ ssize_t qio_channel_writev(QIOChannel *ioc,
size_t niov,
Error **errp)
{
- return qio_channel_writev_full(ioc, iov, niov, NULL, 0, errp);
+ return qio_channel_writev_full(ioc, iov, niov, NULL, 0, 0, errp);
}
@@ -297,7 +313,7 @@ ssize_t qio_channel_write(QIOChannel *ioc,
Error **errp)
{
struct iovec iov = { .iov_base = (char *)buf, .iov_len = buflen };
- return qio_channel_writev_full(ioc, &iov, 1, NULL, 0, errp);
+ return qio_channel_writev_full(ioc, &iov, 1, NULL, 0, 0, errp);
}
@@ -473,6 +489,19 @@ off_t qio_channel_io_seek(QIOChannel *ioc,
return klass->io_seek(ioc, offset, whence, errp);
}
+int qio_channel_flush(QIOChannel *ioc,
+ Error **errp)
+{
+ QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
+
+ if (!klass->io_flush ||
+ !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY)) {
+ return 0;
+ }
+
+ return klass->io_flush(ioc, errp);
+}
+
static void qio_channel_restart_read(void *opaque)
{
diff --git a/migration/rdma.c b/migration/rdma.c
index f5d3bbe7e9..54acd2000e 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -2833,6 +2833,7 @@ static ssize_t qio_channel_rdma_writev(QIOChannel *ioc,
size_t niov,
int *fds,
size_t nfds,
+ int flags,
Error **errp)
{
QIOChannelRDMA *rioc = QIO_CHANNEL_RDMA(ioc);
diff --git a/scsi/pr-manager-helper.c b/scsi/pr-manager-helper.c
index 451c7631b7..3be52a98d5 100644
--- a/scsi/pr-manager-helper.c
+++ b/scsi/pr-manager-helper.c
@@ -77,7 +77,7 @@ static int pr_manager_helper_write(PRManagerHelper *pr_mgr,
iov.iov_base = (void *)buf;
iov.iov_len = sz;
n_written = qio_channel_writev_full(QIO_CHANNEL(pr_mgr->ioc), &iov, 1,
- nfds ? &fd : NULL, nfds, errp);
+ nfds ? &fd : NULL, nfds, 0, errp);
if (n_written <= 0) {
assert(n_written != QIO_CHANNEL_ERR_BLOCK);
diff --git a/tests/unit/test-io-channel-socket.c b/tests/unit/test-io-channel-socket.c
index c49eec1f03..6713886d02 100644
--- a/tests/unit/test-io-channel-socket.c
+++ b/tests/unit/test-io-channel-socket.c
@@ -444,6 +444,7 @@ static void test_io_channel_unix_fd_pass(void)
G_N_ELEMENTS(iosend),
fdsend,
G_N_ELEMENTS(fdsend),
+ 0,
&error_abort);
qio_channel_readv_full(dst,
--
2.35.3

56
SOURCES/kvm-QIOChannelSocket-Add-support-for-MSG_ZEROCOPY-IPV6.patch
Unescape View File

@ -0,0 +1,56 @@
From a6c4aed18a027ce8e107fdf9184e9ea43a86f843 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Thu, 4 Aug 2022 04:10:43 -0300
Subject: [PATCH 8/9] QIOChannelSocket: Add support for MSG_ZEROCOPY + IPV6
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 201: Zero-copy-send fixes + improvements
RH-Commit: [8/8] 6e26ee7c9ebaedb07623313cb0678816867751dd
RH-Bugzilla: 2110203
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
For using MSG_ZEROCOPY, there are two steps:
1 - io_writev() the packet, which enqueues the packet for sending, and
2 - io_flush(), which gets confirmation that all packets got correctly sent
Currently, if MSG_ZEROCOPY is used to send packets over IPV6, no error will
be reported in (1), but it will fail in the first time (2) happens.
This happens because (2) currently checks for cmsg_level & cmsg_type
associated with IPV4 only, before reporting any error.
Add checks for cmsg_level & cmsg_type associated with IPV6, and thus enable
support for MSG_ZEROCOPY + IPV6
Fixes: 2bc58ffc29 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
(cherry picked from commit 5258a7e2c0677d16e9e1d06845f60171adf0b290)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index cf0d67c51b..6010ad7017 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -747,8 +747,8 @@ static int qio_channel_socket_flush(QIOChannel *ioc,
}
cm = CMSG_FIRSTHDR(&msg);
- if (cm->cmsg_level != SOL_IP &&
- cm->cmsg_type != IP_RECVERR) {
+ if (cm->cmsg_level != SOL_IP && cm->cmsg_type != IP_RECVERR &&
+ cm->cmsg_level != SOL_IPV6 && cm->cmsg_type != IPV6_RECVERR) {
error_setg_errno(errp, EPROTOTYPE,
"Wrong cmsg in errqueue");
return -1;
--
2.31.1

65
SOURCES/kvm-QIOChannelSocket-Fix-zero-copy-flush-returning-code-.patch
Unescape View File

@ -0,0 +1,65 @@
From 905cc8032fc63619efb3f0a8c9754b7190bcc43a Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 11 Jul 2022 18:11:11 -0300
Subject: [PATCH 3/9] QIOChannelSocket: Fix zero-copy flush returning code 1
when nothing sent
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 201: Zero-copy-send fixes + improvements
RH-Commit: [3/8] 1ad707702fa26cd4d0fa1870c21f5f26ae93ff97
RH-Bugzilla: 2110203
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
If flush is called when no buffer was sent with MSG_ZEROCOPY, it currently
returns 1. This return code should be used only when Linux fails to use
MSG_ZEROCOPY on a lot of sendmsg().
Fix this by returning early from flush if no sendmsg(...,MSG_ZEROCOPY)
was attempted.
Fixes: 2bc58ffc2926 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20220711211112.18951-2-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 927f93e099c4f9184e60a1bc61624ac2d04d0223)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index df858da924..cf0d67c51b 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -717,12 +717,18 @@ static int qio_channel_socket_flush(QIOChannel *ioc,
struct cmsghdr *cm;
char control[CMSG_SPACE(sizeof(*serr))];
int received;
- int ret = 1;
+ int ret;
+
+ if (sioc->zero_copy_queued == sioc->zero_copy_sent) {
+ return 0;
+ }
msg.msg_control = control;
msg.msg_controllen = sizeof(control);
memset(control, 0, sizeof(control));
+ ret = 1;
+
while (sioc->zero_copy_sent < sioc->zero_copy_queued) {
received = recvmsg(sioc->fd, &msg, MSG_ERRQUEUE);
if (received < 0) {
--
2.31.1

58
SOURCES/kvm-QIOChannelSocket-Fix-zero-copy-send-so-socket-flush-.patch
Unescape View File

@ -0,0 +1,58 @@
From c1fd32d93ae42fcf3c1a25f4d56e669f251087d8 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 20 Jun 2022 02:39:43 -0300
Subject: [PATCH 25/37] QIOChannelSocket: Fix zero-copy send so socket flush
works
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 191: MSG_ZEROCOPY + Multifd @ rhel8.7
RH-Commit: [25/26] 3ede94f3269e21c3ace073ed1a6f24696315bcbb
RH-Bugzilla: 2072049
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Daniel P. Berrangé <berrange@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Somewhere between v6 and v7 the of the zero-copy-send patchset a crucial
part of the flushing mechanism got missing: incrementing zero_copy_queued.
Without that, the flushing interface becomes a no-op, and there is no
guarantee the buffer is really sent.
This can go as bad as causing a corruption in RAM during migration.
Fixes: 2bc58ffc2926 ("QIOChannelSocket: Implement io_writev zero copy flag & io_flush for CONFIG_LINUX")
Reported-by: 徐闯 <xuchuangxclwt@bytedance.com>
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 4f5a09714c983a3471fd12e3c7f3196e95c650c1)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index 7d37b39de7..df858da924 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -612,6 +612,11 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
"Unable to write to socket");
return -1;
}
+
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+ sioc->zero_copy_queued++;
+ }
+
return ret;
}
#else /* WIN32 */
--
2.35.3

249
SOURCES/kvm-QIOChannelSocket-Implement-io_writev-zero-copy-flag-.patch
Unescape View File

@ -0,0 +1,249 @@
From 5fd7af93a06adaddbae719aabbaf912159f4fb28 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Wed, 18 May 2022 02:52:25 -0300
Subject: [PATCH 18/37] QIOChannelSocket: Implement io_writev zero copy flag &
io_flush for CONFIG_LINUX
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 191: MSG_ZEROCOPY + Multifd @ rhel8.7
RH-Commit: [18/26] 6f65c8c879a5df57213b541d58285b65178f8547
RH-Bugzilla: 2072049
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Daniel P. Berrangé <berrange@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
For CONFIG_LINUX, implement the new zero copy flag and the optional callback
io_flush on QIOChannelSocket, but enables it only when MSG_ZEROCOPY
feature is available in the host kernel, which is checked on
qio_channel_socket_connect_sync()
qio_channel_socket_flush() was implemented by counting how many times
sendmsg(...,MSG_ZEROCOPY) was successfully called, and then reading the
socket's error queue, in order to find how many of them finished sending.
Flush will loop until those counters are the same, or until some error occurs.
Notes on using writev() with QIO_CHANNEL_WRITE_FLAG_ZERO_COPY:
1: Buffer
- As MSG_ZEROCOPY tells the kernel to use the same user buffer to avoid copying,
some caution is necessary to avoid overwriting any buffer before it's sent.
If something like this happen, a newer version of the buffer may be sent instead.
- If this is a problem, it's recommended to call qio_channel_flush() before freeing
or re-using the buffer.
2: Locked memory
- When using MSG_ZERCOCOPY, the buffer memory will be locked after queued, and
unlocked after it's sent.
- Depending on the size of each buffer, and how often it's sent, it may require
a larger amount of locked memory than usually available to non-root user.
- If the required amount of locked memory is not available, writev_zero_copy
will return an error, which can abort an operation like migration,
- Because of this, when an user code wants to add zero copy as a feature, it
requires a mechanism to disable it, so it can still be accessible to less
privileged users.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220513062836.965425-4-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 2bc58ffc2926a4efdd03edfb5909861fefc68c3d)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
include/io/channel-socket.h | 2 +
io/channel-socket.c | 116 ++++++++++++++++++++++++++++++++++--
2 files changed, 114 insertions(+), 4 deletions(-)
diff --git a/include/io/channel-socket.h b/include/io/channel-socket.h
index e747e63514..513c428fe4 100644
--- a/include/io/channel-socket.h
+++ b/include/io/channel-socket.h
@@ -47,6 +47,8 @@ struct QIOChannelSocket {
socklen_t localAddrLen;
struct sockaddr_storage remoteAddr;
socklen_t remoteAddrLen;
+ ssize_t zero_copy_queued;
+ ssize_t zero_copy_sent;
};
diff --git a/io/channel-socket.c b/io/channel-socket.c
index bfbd64787e..38a46ba213 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -26,6 +26,14 @@
#include "io/channel-watch.h"
#include "trace.h"
#include "qapi/clone-visitor.h"
+#ifdef CONFIG_LINUX
+#include <linux/errqueue.h>
+#include <sys/socket.h>
+
+#if (defined(MSG_ZEROCOPY) && defined(SO_ZEROCOPY))
+#define QEMU_MSG_ZEROCOPY
+#endif
+#endif
#define SOCKET_MAX_FDS 16
@@ -55,6 +63,8 @@ qio_channel_socket_new(void)
sioc = QIO_CHANNEL_SOCKET(object_new(TYPE_QIO_CHANNEL_SOCKET));
sioc->fd = -1;
+ sioc->zero_copy_queued = 0;
+ sioc->zero_copy_sent = 0;
ioc = QIO_CHANNEL(sioc);
qio_channel_set_feature(ioc, QIO_CHANNEL_FEATURE_SHUTDOWN);
@@ -154,6 +164,16 @@ int qio_channel_socket_connect_sync(QIOChannelSocket *ioc,
return -1;
}
+#ifdef QEMU_MSG_ZEROCOPY
+ int ret, v = 1;
+ ret = setsockopt(fd, SOL_SOCKET, SO_ZEROCOPY, &v, sizeof(v));
+ if (ret == 0) {
+ /* Zero copy available on host */
+ qio_channel_set_feature(QIO_CHANNEL(ioc),
+ QIO_CHANNEL_FEATURE_WRITE_ZERO_COPY);
+ }
+#endif
+
return 0;
}
@@ -534,6 +554,7 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
char control[CMSG_SPACE(sizeof(int) * SOCKET_MAX_FDS)];
size_t fdsize = sizeof(int) * nfds;
struct cmsghdr *cmsg;
+ int sflags = 0;
memset(control, 0, CMSG_SPACE(sizeof(int) * SOCKET_MAX_FDS));
@@ -558,15 +579,31 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
memcpy(CMSG_DATA(cmsg), fds, fdsize);
}
+#ifdef QEMU_MSG_ZEROCOPY
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+ sflags = MSG_ZEROCOPY;
+ }
+#endif
+
retry:
- ret = sendmsg(sioc->fd, &msg, 0);
+ ret = sendmsg(sioc->fd, &msg, sflags);
if (ret <= 0) {
- if (errno == EAGAIN) {
+ switch (errno) {
+ case EAGAIN:
return QIO_CHANNEL_ERR_BLOCK;
- }
- if (errno == EINTR) {
+ case EINTR:
goto retry;
+#ifdef QEMU_MSG_ZEROCOPY
+ case ENOBUFS:
+ if (sflags & MSG_ZEROCOPY) {
+ error_setg_errno(errp, errno,
+ "Process can't lock enough memory for using MSG_ZEROCOPY");
+ return -1;
+ }
+ break;
+#endif
}
+
error_setg_errno(errp, errno,
"Unable to write to socket");
return -1;
@@ -660,6 +697,74 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
}
#endif /* WIN32 */
+
+#ifdef QEMU_MSG_ZEROCOPY
+static int qio_channel_socket_flush(QIOChannel *ioc,
+ Error **errp)
+{
+ QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
+ struct msghdr msg = {};
+ struct sock_extended_err *serr;
+ struct cmsghdr *cm;
+ char control[CMSG_SPACE(sizeof(*serr))];
+ int received;
+ int ret = 1;
+
+ msg.msg_control = control;
+ msg.msg_controllen = sizeof(control);
+ memset(control, 0, sizeof(control));
+
+ while (sioc->zero_copy_sent < sioc->zero_copy_queued) {
+ received = recvmsg(sioc->fd, &msg, MSG_ERRQUEUE);
+ if (received < 0) {
+ switch (errno) {
+ case EAGAIN:
+ /* Nothing on errqueue, wait until something is available */
+ qio_channel_wait(ioc, G_IO_ERR);
+ continue;
+ case EINTR:
+ continue;
+ default:
+ error_setg_errno(errp, errno,
+ "Unable to read errqueue");
+ return -1;
+ }
+ }
+
+ cm = CMSG_FIRSTHDR(&msg);
+ if (cm->cmsg_level != SOL_IP &&
+ cm->cmsg_type != IP_RECVERR) {
+ error_setg_errno(errp, EPROTOTYPE,
+ "Wrong cmsg in errqueue");
+ return -1;
+ }
+
+ serr = (void *) CMSG_DATA(cm);
+ if (serr->ee_errno != SO_EE_ORIGIN_NONE) {
+ error_setg_errno(errp, serr->ee_errno,
+ "Error on socket");
+ return -1;
+ }
+ if (serr->ee_origin != SO_EE_ORIGIN_ZEROCOPY) {
+ error_setg_errno(errp, serr->ee_origin,
+ "Error not from zero copy");
+ return -1;
+ }
+
+ /* No errors, count successfully finished sendmsg()*/
+ sioc->zero_copy_sent += serr->ee_data - serr->ee_info + 1;
+
+ /* If any sendmsg() succeeded using zero copy, return 0 at the end */
+ if (serr->ee_code != SO_EE_CODE_ZEROCOPY_COPIED) {
+ ret = 0;
+ }
+ }
+
+ return ret;
+}
+
+#endif /* QEMU_MSG_ZEROCOPY */
+
static int
qio_channel_socket_set_blocking(QIOChannel *ioc,
bool enabled,
@@ -789,6 +894,9 @@ static void qio_channel_socket_class_init(ObjectClass *klass,
ioc_klass->io_set_delay = qio_channel_socket_set_delay;
ioc_klass->io_create_watch = qio_channel_socket_create_watch;
ioc_klass->io_set_aio_fd_handler = qio_channel_socket_set_aio_fd_handler;
+#ifdef QEMU_MSG_ZEROCOPY
+ ioc_klass->io_flush = qio_channel_socket_flush;
+#endif
}
static const TypeInfo qio_channel_socket_info = {
--
2.35.3

82
SOURCES/kvm-QIOChannelSocket-Introduce-assert-and-reduce-ifdefs-.patch
Unescape View File

@ -0,0 +1,82 @@
From cbfaf86331c2b2e01a2083303b7554672bf991b7 Mon Sep 17 00:00:00 2001
From: Leonardo Bras <leobras@redhat.com>
Date: Mon, 20 Jun 2022 02:39:42 -0300
Subject: [PATCH 24/37] QIOChannelSocket: Introduce assert and reduce ifdefs to
improve readability
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Leonardo Brás <leobras@redhat.com>
RH-MergeRequest: 191: MSG_ZEROCOPY + Multifd @ rhel8.7
RH-Commit: [24/26] b50e2e65307149f247155a7f7a032dc99e57718d
RH-Bugzilla: 2072049
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Daniel P. Berrangé <berrange@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
During implementation of MSG_ZEROCOPY feature, a lot of #ifdefs were
introduced, particularly at qio_channel_socket_writev().
Rewrite some of those changes so it's easier to read.
Also, introduce an assert to help detect incorrect zero-copy usage is when
it's disabled on build.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Fixed up thinko'd g_assert_unreachable->g_assert_not_reached
(cherry picked from commit 803ca43e4c7fcf32f9f68c118301ccd0c83ece3f)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
---
io/channel-socket.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/io/channel-socket.c b/io/channel-socket.c
index 38a46ba213..7d37b39de7 100644
--- a/io/channel-socket.c
+++ b/io/channel-socket.c
@@ -579,11 +579,17 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
memcpy(CMSG_DATA(cmsg), fds, fdsize);
}
-#ifdef QEMU_MSG_ZEROCOPY
if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
+#ifdef QEMU_MSG_ZEROCOPY
sflags = MSG_ZEROCOPY;
- }
+#else
+ /*
+ * We expect QIOChannel class entry point to have
+ * blocked this code path already
+ */
+ g_assert_not_reached();
#endif
+ }
retry:
ret = sendmsg(sioc->fd, &msg, sflags);
@@ -593,15 +599,13 @@ static ssize_t qio_channel_socket_writev(QIOChannel *ioc,
return QIO_CHANNEL_ERR_BLOCK;
case EINTR:
goto retry;
-#ifdef QEMU_MSG_ZEROCOPY
case ENOBUFS:
- if (sflags & MSG_ZEROCOPY) {
+ if (flags & QIO_CHANNEL_WRITE_FLAG_ZERO_COPY) {
error_setg_errno(errp, errno,
"Process can't lock enough memory for using MSG_ZEROCOPY");
return -1;
}
break;
-#endif
}
error_setg_errno(errp, errno,
--
2.35.3

107
SOURCES/kvm-RHEL-disable-seqpacket-for-vhost-vsock-device-in-rhe.patch
Unescape View File

@ -0,0 +1,107 @@
From e0e4f01c6f4fb5881960f72ae4e80951b711131e Mon Sep 17 00:00:00 2001
From: Stefano Garzarella <sgarzare@redhat.com>
Date: Thu, 24 Mar 2022 16:04:57 +0100
Subject: [PATCH 1/5] RHEL: disable "seqpacket" for "vhost-vsock-device" in
rhel8.6.0
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Stefano Garzarella <sgarzare@redhat.com>
RH-MergeRequest: 136: RHEL: disable "seqpacket" for "vhost-vsock-device" in rhel8.6.0 [rhel-8.7.0]
RH-Commit: [1/1] d82ea09e123679521503689f7d9af1c03dc71bfc
RH-Bugzilla: 2068202
RH-Acked-by: Jason Wang <None>
RH-Acked-by: Eugenio Pérez <None>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
vhost-vsock device in RHEL 8 kernels doesn't support seqpacket.
To avoid problems when migrating a VM from RHEL 9 host, we need to
disable it in rhel8-* machine types.
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
hw/core/machine.c | 10 ++++++++++
hw/i386/pc_piix.c | 2 ++
hw/i386/pc_q35.c | 2 ++
hw/s390x/s390-virtio-ccw.c | 1 +
include/hw/boards.h | 3 +++
5 files changed, 18 insertions(+)
diff --git a/hw/core/machine.c b/hw/core/machine.c
index 024b025fc2..76fcabec7a 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -37,6 +37,16 @@
#include "hw/virtio/virtio.h"
#include "hw/virtio/virtio-pci.h"
+GlobalProperty hw_compat_rhel_8_6[] = {
+ /* hw_compat_rhel_8_6 bz 2068202 */
+ /*
+ * vhost-vsock device in RHEL 8 kernels doesn't support seqpacket, so
+ * we need do disable it downstream on the latest hw_compat_rhel_8.
+ */
+ { "vhost-vsock-device", "seqpacket", "off" },
+};
+const size_t hw_compat_rhel_8_6_len = G_N_ELEMENTS(hw_compat_rhel_8_6);
+
/*
* Mostly the same as hw_compat_6_0 and hw_compat_6_1
*/
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index f03a8f0db8..ab6d03e07a 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -998,6 +998,8 @@ static void pc_machine_rhel760_options(MachineClass *m)
pcmc->kvmclock_create_always = false;
/* From pc_i440fx_5_1_machine_options() */
pcmc->pci_root_uid = 1;
+ compat_props_add(m->compat_props, hw_compat_rhel_8_6,
+ hw_compat_rhel_8_6_len);
compat_props_add(m->compat_props, hw_compat_rhel_8_5,
hw_compat_rhel_8_5_len);
compat_props_add(m->compat_props, pc_rhel_8_5_compat,
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index 5559261d9e..882fe7a68d 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -658,6 +658,8 @@ static void pc_q35_machine_rhel860_options(MachineClass *m)
m->desc = "RHEL-8.6.0 PC (Q35 + ICH9, 2009)";
pcmc->smbios_stream_product = "RHEL-AV";
pcmc->smbios_stream_version = "8.6.0";
+ compat_props_add(m->compat_props, hw_compat_rhel_8_6,
+ hw_compat_rhel_8_6_len);
}
DEFINE_PC_MACHINE(q35_rhel860, "pc-q35-rhel8.6.0", pc_q35_init_rhel860,
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index 9795eb9406..bec270598b 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -1109,6 +1109,7 @@ static void ccw_machine_rhel860_instance_options(MachineState *machine)
static void ccw_machine_rhel860_class_options(MachineClass *mc)
{
+ compat_props_add(mc->compat_props, hw_compat_rhel_8_6, hw_compat_rhel_8_6_len);
}
DEFINE_CCW_MACHINE(rhel860, "rhel8.6.0", true);
diff --git a/include/hw/boards.h b/include/hw/boards.h
index 04e8759815..4ddb798144 100644
--- a/include/hw/boards.h
+++ b/include/hw/boards.h
@@ -443,6 +443,9 @@ extern const size_t hw_compat_2_2_len;
extern GlobalProperty hw_compat_2_1[];
extern const size_t hw_compat_2_1_len;
+extern GlobalProperty hw_compat_rhel_8_6[];
+extern const size_t hw_compat_rhel_8_6_len;
+
extern GlobalProperty hw_compat_rhel_8_5[];
extern const size_t hw_compat_rhel_8_5_len;
--
2.27.0

93
SOURCES/kvm-Revert-redhat-Add-hw_compat_4_2_extra-and-apply-to-u.patch
Unescape View File

@ -0,0 +1,93 @@
From e626dc16d130c724c400b99a93daad0a9abeae59 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Tue, 22 Mar 2022 19:23:36 -0400
Subject: [PATCH 01/18] Revert "redhat: Add hw_compat_4_2_extra and apply to
upstream machines"
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 131: Revert "redhat: Add hw_compat_4_2_extra and apply to upstream machines"
RH-Commit: [1/3] 47b7d9e5062f5e215d5ed1a3ecdc1a87ac3fa630 (jmaloy/qemu-kvm)
RH-Bugzilla: 2062613
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
BZ: https://bugzilla.redhat.com/2062613
UPSTREAM: no
BREW: https://brewweb.engineering.redhat.com/brew/taskinfo?taskID=44038000
commit dc2e9ec1e014950c7918e23a3e9b0096b34a4a92
Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
Date: Wed Mar 9 10:31:53 2022 +0000
Revert "redhat: Add hw_compat_4_2_extra and apply to upstream machines"
This reverts commit 66882f9a3230246409f3918424aca26add5c034a.
We no longer need these compat machines it was added for.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit dc2e9ec1e014950c7918e23a3e9b0096b34a4a92)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/i386/pc.c | 12 ------------
hw/i386/pc_piix.c | 6 ------
include/hw/i386/pc.h | 3 ---
3 files changed, 21 deletions(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 4c08a1971c..357257349b 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -670,18 +670,6 @@ GlobalProperty pc_rhel_7_0_compat[] = {
};
const size_t pc_rhel_7_0_compat_len = G_N_ELEMENTS(pc_rhel_7_0_compat);
-/*
- * RHEL: These properties only apply to the RHEL exported machine types
- * pc-4.2/2.11 for the purpose to have a limited upstream machines support
- * which can be migrated to RHEL. Let's avoid touching hw_compat_4_2 directly
- * so that we can have some isolation against the upstream code.
- */
-GlobalProperty hw_compat_4_2_extra[] = {
- /* By default enlarge the default virtio-net-pci ROM to 512KB. */
- { "virtio-net-pci", "romsize", "0x80000" },
-};
-const size_t hw_compat_4_2_extra_len = G_N_ELEMENTS(hw_compat_4_2_extra);
-
GSIState *pc_gsi_create(qemu_irq **irqs, bool pci_enabled)
{
GSIState *s;
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index c30057c443..7b7076cbc7 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -531,12 +531,6 @@ static void pc_i440fx_4_2_machine_options(MachineClass *m)
* supported by RHEL, even if exported.
*/
m->deprecation_reason = "Not supported by RHEL";
- /*
- * RHEL: Specific compat properties to have limited support for upstream
- * machines exported.
- */
- compat_props_add(m->compat_props, hw_compat_4_2_extra,
- hw_compat_4_2_extra_len);
}
/* RHEL: Export pc-4.2 */
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 9e8bfb69f8..4a593acb50 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -325,9 +325,6 @@ extern const size_t pc_rhel_7_1_compat_len;
extern GlobalProperty pc_rhel_7_0_compat[];
extern const size_t pc_rhel_7_0_compat_len;
-extern GlobalProperty hw_compat_4_2_extra[];
-extern const size_t hw_compat_4_2_extra_len;
-
/* Helper for setting model-id for CPU models that changed model-id
* depending on QEMU versions up to QEMU 2.4.
*/
--
2.27.0

128
SOURCES/kvm-Revert-redhat-Add-some-devices-for-exporting-upstrea.patch
Unescape View File

@ -0,0 +1,128 @@
From 96edd15df257f1d1496397a6fac24b4316570d7e Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Thu, 14 Apr 2022 16:45:30 -0400
Subject: [PATCH 1/3] Revert redhat: Add some devices for exporting upstream
machine types
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 156: Revert redhat: Add some devices for exporting upstream machine types
RH-Commit: [1/1] f25d0da3a181136917ead82f5a5c59efe3fa445a (jmaloy/qemu-kvm)
RH-Bugzilla: 2065043
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2065043
Upstream: no
Manual revert of commit 70d3924521c9bfd912bcf1a1fc76f49eb377de46, since
the directory structure looks different from rhel-av-8.4.0.z where
this commit is taken from. Besides, x86_64-softmmu.mak looks totally
different and should not be affected by this reversal.
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
configs/devices/x86_64-softmmu/x86_64-rh-devices.mak | 1 -
.../devices/x86_64-softmmu/x86_64-upstream-devices.mak | 4 ----
hw/char/parallel.c | 9 ---------
hw/i386/pc_piix.c | 2 +-
hw/i386/pc_q35.c | 2 +-
hw/timer/hpet.c | 8 --------
6 files changed, 2 insertions(+), 24 deletions(-)
delete mode 100644 configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak
diff --git a/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak b/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
index fdbbdf9742..31ce08edab 100644
--- a/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
+++ b/configs/devices/x86_64-softmmu/x86_64-rh-devices.mak
@@ -1,5 +1,4 @@
include ../rh-virtio.mak
-include x86_64-upstream-devices.mak
CONFIG_AC97=y
CONFIG_ACPI=y
diff --git a/configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak b/configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak
deleted file mode 100644
index 2cd20f54d2..0000000000
--- a/configs/devices/x86_64-softmmu/x86_64-upstream-devices.mak
+++ /dev/null
@@ -1,4 +0,0 @@
-# We need "isa-parallel"
-CONFIG_PARALLEL=y
-# We need "hpet"
-CONFIG_HPET=y
diff --git a/hw/char/parallel.c b/hw/char/parallel.c
index e5f108211b..b45e67bfbb 100644
--- a/hw/char/parallel.c
+++ b/hw/char/parallel.c
@@ -29,7 +29,6 @@
#include "chardev/char-parallel.h"
#include "chardev/char-fe.h"
#include "hw/acpi/aml-build.h"
-#include "hw/boards.h"
#include "hw/irq.h"
#include "hw/isa/isa.h"
#include "hw/qdev-properties.h"
@@ -535,14 +534,6 @@ static void parallel_isa_realizefn(DeviceState *dev, Error **errp)
int base;
uint8_t dummy;
- /* Restricted for Red Hat Enterprise Linux */
- MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
- if (strstr(mc->name, "rhel")) {
- error_setg(errp, "Device %s is not supported with machine type %s",
- object_get_typename(OBJECT(dev)), mc->name);
- return;
- }
-
if (!qemu_chr_fe_backend_connected(&s->chr)) {
error_setg(errp, "Can't create parallel device, empty char device");
return;
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index ab6d03e07a..5f101c8748 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -966,7 +966,7 @@ static void pc_machine_rhel7_options(MachineClass *m)
{
PCMachineClass *pcmc = PC_MACHINE_CLASS(m);
m->family = "pc_piix_Y";
- m->default_machine_opts = "firmware=bios-256k.bin,hpet=off";
+ m->default_machine_opts = "firmware=bios-256k.bin";
pcmc->default_nic_model = "e1000";
pcmc->pci_root_uid = 0;
m->default_display = "std";
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index 882fe7a68d..73b0d0d317 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -633,7 +633,7 @@ static void pc_q35_machine_rhel_options(MachineClass *m)
pcmc->pci_root_uid = 0;
m->family = "pc_q35_Z";
m->units_per_default_bus = 1;
- m->default_machine_opts = "firmware=bios-256k.bin,hpet=off";
+ m->default_machine_opts = "firmware=bios-256k.bin";
m->default_display = "std";
m->no_floppy = 1;
m->no_parallel = 1;
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index 202e032524..9520471be2 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -733,14 +733,6 @@ static void hpet_realize(DeviceState *dev, Error **errp)
int i;
HPETTimer *timer;
- /* Restricted for Red Hat Enterprise Linux */
- MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
- if (strstr(mc->name, "rhel")) {
- error_setg(errp, "Device %s is not supported with machine type %s",
- object_get_typename(OBJECT(dev)), mc->name);
- return;
- }
-
if (!s->intcap) {
warn_report("Hpet's intcap not initialized");
}
--
2.35.1

53
SOURCES/kvm-Revert-redhat-Enable-FDC-device-for-upstream-machine.patch
Unescape View File

@ -0,0 +1,53 @@
From 5bf8f1d69fea1225e927fbb3efe549a2a9d47d92 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Tue, 22 Mar 2022 19:23:36 -0400
Subject: [PATCH 02/18] Revert "redhat: Enable FDC device for upstream machines
too"
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 131: Revert "redhat: Add hw_compat_4_2_extra and apply to upstream machines"
RH-Commit: [2/3] 4e3c945e3de9bb9d9a6d24115f0719168c9669fe (jmaloy/qemu-kvm)
RH-Bugzilla: 2062613
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
BZ: https://bugzilla.redhat.com/2062613
UPSTREAM: no
BREW: https://brewweb.engineering.redhat.com/brew/taskinfo?taskID=44038000
commit 597cb6ca1da4a3eea77c1e4928f55203a1d5c70c
Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
Date: Wed Mar 9 10:32:39 2022 +0000
Revert "redhat: Enable FDC device for upstream machines too"
This reverts commit c4d1aa8bf21fe98da94a9cff30b7c25bed12c17f.
We no longer need these compat machines it was added for.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit 597cb6ca1da4a3eea77c1e4928f55203a1d5c70c)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/block/fdc.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/hw/block/fdc.c b/hw/block/fdc.c
index 63042ef030..97fa6de423 100644
--- a/hw/block/fdc.c
+++ b/hw/block/fdc.c
@@ -2341,10 +2341,7 @@ void fdctrl_realize_common(DeviceState *dev, FDCtrl *fdctrl, Error **errp)
/* Restricted for Red Hat Enterprise Linux: */
MachineClass *mc = MACHINE_GET_CLASS(qdev_get_machine());
- if (!strstr(mc->name, "-rhel7.") &&
- /* Exported two upstream machine types allows FDC too */
- strcmp(mc->name, "pc-i440fx-4.2") &&
- strcmp(mc->name, "pc-i440fx-2.11")) {
+ if (!strstr(mc->name, "-rhel7.")) {
error_setg(errp, "Device %s is not supported with machine type %s",
object_get_typename(OBJECT(dev)), mc->name);
return;
--
2.27.0

191
SOURCES/kvm-Revert-redhat-Expose-upstream-machines-pc-4.2-and-pc.patch
Unescape View File

@ -0,0 +1,191 @@
From ee3cae3bb349469edcf725a1c5161521e95dcb9f Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Tue, 22 Mar 2022 19:23:36 -0400
Subject: [PATCH 03/18] Revert "redhat: Expose upstream machines pc-4.2 and
pc-2.11"
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 131: Revert "redhat: Add hw_compat_4_2_extra and apply to upstream machines"
RH-Commit: [3/3] 35cee68034580f81b3aa916921eecd2fdfa7dd15 (jmaloy/qemu-kvm)
RH-Bugzilla: 2062613
RH-Acked-by: Peter Xu <peterx@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
BZ: https://bugzilla.redhat.com/2062613
UPSTREAM: no
BREW: https://brewweb.engineering.redhat.com/brew/taskinfo?taskID=44038000
commit f3b50d6d4ae0be9e64aafe6a15f5423bab4899e9
Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
Date: Wed Mar 9 10:34:58 2022 +0000
Revert "redhat: Expose upstream machines pc-4.2 and pc-2.11"
This reverts commit 618e2424edba499d52cd26cf8363bc2dd85ef149.
We no longer need these compat machines.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
(cherry picked from commit f3b50d6d4ae0be9e64aafe6a15f5423bab4899e9)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/i386/pc_piix.c | 37 -------------------------------------
1 file changed, 37 deletions(-)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 7b7076cbc7..f03a8f0db8 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -315,14 +315,6 @@ static void pc_init1(MachineState *machine,
* hw_compat_*, pc_compat_*, or * pc_*_machine_options().
*/
-/*
- * NOTE! Not all the upstream machine types are disabled for RHEL. For
- * providing a very limited support for upstream machine types, pc machines
- * 2.11 and 4.2 are exposed explicitly. This will make the below "#if" macros
- * a bit messed up, but please read this comment first so that we can have a
- * rough understanding of what we're going to do.
- */
-
#if 0 /* Disabled for Red Hat Enterprise Linux */
static void pc_compat_2_3_fn(MachineState *machine)
{
@@ -399,8 +391,6 @@ static void pc_xen_hvm_init(MachineState *machine)
}
#endif
-#endif /* Disabled for Red Hat Enterprise Linux */
-
#define DEFINE_I440FX_MACHINE(suffix, name, compatfn, optionfn) \
static void pc_init_##suffix(MachineState *machine) \
{ \
@@ -465,10 +455,8 @@ static void pc_i440fx_6_0_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_6_0, pc_compat_6_0_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v6_0, "pc-i440fx-6.0", NULL,
pc_i440fx_6_0_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_5_2_machine_options(MachineClass *m)
{
@@ -479,10 +467,8 @@ static void pc_i440fx_5_2_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_5_2, pc_compat_5_2_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v5_2, "pc-i440fx-5.2", NULL,
pc_i440fx_5_2_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_5_1_machine_options(MachineClass *m)
{
@@ -497,10 +483,8 @@ static void pc_i440fx_5_1_machine_options(MachineClass *m)
pcmc->pci_root_uid = 1;
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v5_1, "pc-i440fx-5.1", NULL,
pc_i440fx_5_1_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_5_0_machine_options(MachineClass *m)
{
@@ -513,10 +497,8 @@ static void pc_i440fx_5_0_machine_options(MachineClass *m)
m->auto_enable_numa_with_memdev = false;
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v5_0, "pc-i440fx-5.0", NULL,
pc_i440fx_5_0_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_4_2_machine_options(MachineClass *m)
{
@@ -525,15 +507,8 @@ static void pc_i440fx_4_2_machine_options(MachineClass *m)
m->is_default = false;
compat_props_add(m->compat_props, hw_compat_4_2, hw_compat_4_2_len);
compat_props_add(m->compat_props, pc_compat_4_2, pc_compat_4_2_len);
-
- /*
- * RHEL: Mark all upstream machines as deprecated because they're not
- * supported by RHEL, even if exported.
- */
- m->deprecation_reason = "Not supported by RHEL";
}
-/* RHEL: Export pc-4.2 */
DEFINE_I440FX_MACHINE(v4_2, "pc-i440fx-4.2", NULL,
pc_i440fx_4_2_machine_options);
@@ -546,10 +521,8 @@ static void pc_i440fx_4_1_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_4_1, pc_compat_4_1_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v4_1, "pc-i440fx-4.1", NULL,
pc_i440fx_4_1_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_4_0_machine_options(MachineClass *m)
{
@@ -562,10 +535,8 @@ static void pc_i440fx_4_0_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_4_0, pc_compat_4_0_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v4_0, "pc-i440fx-4.0", NULL,
pc_i440fx_4_0_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_3_1_machine_options(MachineClass *m)
{
@@ -581,10 +552,8 @@ static void pc_i440fx_3_1_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_3_1, pc_compat_3_1_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v3_1, "pc-i440fx-3.1", NULL,
pc_i440fx_3_1_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_3_0_machine_options(MachineClass *m)
{
@@ -593,10 +562,8 @@ static void pc_i440fx_3_0_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_3_0, pc_compat_3_0_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v3_0, "pc-i440fx-3.0", NULL,
pc_i440fx_3_0_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_2_12_machine_options(MachineClass *m)
{
@@ -605,10 +572,8 @@ static void pc_i440fx_2_12_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_2_12, pc_compat_2_12_len);
}
-#if 0 /* Disabled for Red Hat Enterprise Linux */
DEFINE_I440FX_MACHINE(v2_12, "pc-i440fx-2.12", NULL,
pc_i440fx_2_12_machine_options);
-#endif /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_2_11_machine_options(MachineClass *m)
{
@@ -617,11 +582,9 @@ static void pc_i440fx_2_11_machine_options(MachineClass *m)
compat_props_add(m->compat_props, pc_compat_2_11, pc_compat_2_11_len);
}
-/* RHEL: Export pc-2.11 */
DEFINE_I440FX_MACHINE(v2_11, "pc-i440fx-2.11", NULL,
pc_i440fx_2_11_machine_options);
-#if 0 /* Disabled for Red Hat Enterprise Linux */
static void pc_i440fx_2_10_machine_options(MachineClass *m)
{
pc_i440fx_2_11_machine_options(m);
--
2.27.0

171
SOURCES/kvm-Update-linux-headers-to-v6.0-rc4.patch
Unescape View File

@ -0,0 +1,171 @@
From 10fc28b61a6fba1e6dc44fd544cf31c7f313c622 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?C=C3=A9dric=20Le=20Goater?= <clg@redhat.com>
Date: Fri, 28 Oct 2022 17:48:00 +0100
Subject: [PATCH 05/42] Update linux headers to v6.0-rc4
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [5/41] ca55f497d1bf1e72179330f8f613781bf999d898
Based on upstream commit d525f73f9186a5bc641b8caf0b2c9bb94e5aa963
("Update linux headers to v6.0-rc4"), but this is focusing only on the
ZPCI and protected dump changes.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
linux-headers/linux/kvm.h | 87 +++++++++++++++++++++++++++++++++
linux-headers/linux/vfio_zdev.h | 7 +++
2 files changed, 94 insertions(+)
diff --git a/linux-headers/linux/kvm.h b/linux-headers/linux/kvm.h
index 0d05d02ee4..c65930288c 100644
--- a/linux-headers/linux/kvm.h
+++ b/linux-headers/linux/kvm.h
@@ -1150,6 +1150,9 @@ struct kvm_ppc_resize_hpt {
#define KVM_CAP_DISABLE_QUIRKS2 213
/* #define KVM_CAP_VM_TSC_CONTROL 214 */
#define KVM_CAP_SYSTEM_EVENT_DATA 215
+#define KVM_CAP_S390_PROTECTED_DUMP 217
+#define KVM_CAP_S390_ZPCI_OP 221
+#define KVM_CAP_S390_CPU_TOPOLOGY 222
#ifdef KVM_CAP_IRQ_ROUTING
@@ -1651,6 +1654,55 @@ struct kvm_s390_pv_unp {
__u64 tweak;
};
+enum pv_cmd_dmp_id {
+ KVM_PV_DUMP_INIT,
+ KVM_PV_DUMP_CONFIG_STOR_STATE,
+ KVM_PV_DUMP_COMPLETE,
+ KVM_PV_DUMP_CPU,
+};
+
+struct kvm_s390_pv_dmp {
+ __u64 subcmd;
+ __u64 buff_addr;
+ __u64 buff_len;
+ __u64 gaddr; /* For dump storage state */
+ __u64 reserved[4];
+};
+
+enum pv_cmd_info_id {
+ KVM_PV_INFO_VM,
+ KVM_PV_INFO_DUMP,
+};
+
+struct kvm_s390_pv_info_dump {
+ __u64 dump_cpu_buffer_len;
+ __u64 dump_config_mem_buffer_per_1m;
+ __u64 dump_config_finalize_len;
+};
+
+struct kvm_s390_pv_info_vm {
+ __u64 inst_calls_list[4];
+ __u64 max_cpus;
+ __u64 max_guests;
+ __u64 max_guest_addr;
+ __u64 feature_indication;
+};
+
+struct kvm_s390_pv_info_header {
+ __u32 id;
+ __u32 len_max;
+ __u32 len_written;
+ __u32 reserved;
+};
+
+struct kvm_s390_pv_info {
+ struct kvm_s390_pv_info_header header;
+ union {
+ struct kvm_s390_pv_info_dump dump;
+ struct kvm_s390_pv_info_vm vm;
+ };
+};
+
enum pv_cmd_id {
KVM_PV_ENABLE,
KVM_PV_DISABLE,
@@ -1659,6 +1711,8 @@ enum pv_cmd_id {
KVM_PV_VERIFY,
KVM_PV_PREP_RESET,
KVM_PV_UNSHARE_ALL,
+ KVM_PV_INFO,
+ KVM_PV_DUMP,
};
struct kvm_pv_cmd {
@@ -2066,4 +2120,37 @@ struct kvm_stats_desc {
/* Available with KVM_CAP_XSAVE2 */
#define KVM_GET_XSAVE2 _IOR(KVMIO, 0xcf, struct kvm_xsave)
+/* Available with KVM_CAP_S390_PROTECTED_DUMP */
+#define KVM_S390_PV_CPU_COMMAND _IOWR(KVMIO, 0xd0, struct kvm_pv_cmd)
+
+/* Available with KVM_CAP_S390_ZPCI_OP */
+#define KVM_S390_ZPCI_OP _IOW(KVMIO, 0xd1, struct kvm_s390_zpci_op)
+
+struct kvm_s390_zpci_op {
+ /* in */
+ __u32 fh; /* target device */
+ __u8 op; /* operation to perform */
+ __u8 pad[3];
+ union {
+ /* for KVM_S390_ZPCIOP_REG_AEN */
+ struct {
+ __u64 ibv; /* Guest addr of interrupt bit vector */
+ __u64 sb; /* Guest addr of summary bit */
+ __u32 flags;
+ __u32 noi; /* Number of interrupts */
+ __u8 isc; /* Guest interrupt subclass */
+ __u8 sbo; /* Offset of guest summary bit vector */
+ __u16 pad;
+ } reg_aen;
+ __u64 reserved[8];
+ } u;
+};
+
+/* types for kvm_s390_zpci_op->op */
+#define KVM_S390_ZPCIOP_REG_AEN 0
+#define KVM_S390_ZPCIOP_DEREG_AEN 1
+
+/* flags for kvm_s390_zpci_op->u.reg_aen.flags */
+#define KVM_S390_ZPCIOP_REGAEN_HOST (1 << 0)
+
#endif /* __LINUX_KVM_H */
diff --git a/linux-headers/linux/vfio_zdev.h b/linux-headers/linux/vfio_zdev.h
index b4309397b6..77f2aff1f2 100644
--- a/linux-headers/linux/vfio_zdev.h
+++ b/linux-headers/linux/vfio_zdev.h
@@ -29,6 +29,9 @@ struct vfio_device_info_cap_zpci_base {
__u16 fmb_length; /* Measurement Block Length (in bytes) */
__u8 pft; /* PCI Function Type */
__u8 gid; /* PCI function group ID */
+ /* End of version 1 */
+ __u32 fh; /* PCI function handle */
+ /* End of version 2 */
};
/**
@@ -47,6 +50,10 @@ struct vfio_device_info_cap_zpci_group {
__u16 noi; /* Maximum number of MSIs */
__u16 maxstbl; /* Maximum Store Block Length */
__u8 version; /* Supported PCI Version */
+ /* End of version 1 */
+ __u8 reserved;
+ __u16 imaxstbl; /* Maximum Interpreted Store Block Length */
+ /* End of version 2 */
};
/**
--
2.37.3

349
SOURCES/kvm-accel-introduce-accelerator-blocker-API.patch
Unescape View File

@ -0,0 +1,349 @@
From a5e7bb1f7a88efb5574266a76e80fd7604d19921 Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Mon, 16 Jan 2023 07:49:59 -0500
Subject: [PATCH 04/11] accel: introduce accelerator blocker API
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 247: accel: introduce accelerator blocker API
RH-Bugzilla: 2161188
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Vitaly Kuznetsov <vkuznets@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Commit: [1/3] 9d3d7f9554974a79042c915763288cce07aef135
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2161188
commit bd688fc93120fb3e28aa70e3dfdf567ccc1e0bc1
Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Fri Nov 11 10:47:56 2022 -0500
accel: introduce accelerator blocker API
This API allows the accelerators to prevent vcpus from issuing
new ioctls while execting a critical section marked with the
accel_ioctl_inhibit_begin/end functions.
Note that all functions submitting ioctls must mark where the
ioctl is being called with accel_{cpu_}ioctl_begin/end().
This API requires the caller to always hold the BQL.
API documentation is in sysemu/accel-blocker.h
Internally, it uses a QemuLockCnt together with a per-CPU QemuLockCnt
(to minimize cache line bouncing) to keep avoid that new ioctls
run when the critical section starts, and a QemuEvent to wait
that all running ioctls finish.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221111154758.1372674-2-eesposit@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Conflicts:
util/meson.build: files are missing in rhel 8.8.0
namely int128.c, memalign.c and interval-tree.c
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
accel/accel-blocker.c | 154 +++++++++++++++++++++++++++++++++
accel/meson.build | 2 +-
hw/core/cpu-common.c | 2 +
include/hw/core/cpu.h | 3 +
include/sysemu/accel-blocker.h | 56 ++++++++++++
util/meson.build | 2 +-
6 files changed, 217 insertions(+), 2 deletions(-)
create mode 100644 accel/accel-blocker.c
create mode 100644 include/sysemu/accel-blocker.h
diff --git a/accel/accel-blocker.c b/accel/accel-blocker.c
new file mode 100644
index 0000000000..1e7f423462
--- /dev/null
+++ b/accel/accel-blocker.c
@@ -0,0 +1,154 @@
+/*
+ * Lock to inhibit accelerator ioctls
+ *
+ * Copyright (c) 2022 Red Hat Inc.
+ *
+ * Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/thread.h"
+#include "qemu/main-loop.h"
+#include "hw/core/cpu.h"
+#include "sysemu/accel-blocker.h"
+
+static QemuLockCnt accel_in_ioctl_lock;
+static QemuEvent accel_in_ioctl_event;
+
+void accel_blocker_init(void)
+{
+ qemu_lockcnt_init(&accel_in_ioctl_lock);
+ qemu_event_init(&accel_in_ioctl_event, false);
+}
+
+void accel_ioctl_begin(void)
+{
+ if (likely(qemu_mutex_iothread_locked())) {
+ return;
+ }
+
+ /* block if lock is taken in kvm_ioctl_inhibit_begin() */
+ qemu_lockcnt_inc(&accel_in_ioctl_lock);
+}
+
+void accel_ioctl_end(void)
+{
+ if (likely(qemu_mutex_iothread_locked())) {
+ return;
+ }
+
+ qemu_lockcnt_dec(&accel_in_ioctl_lock);
+ /* change event to SET. If event was BUSY, wake up all waiters */
+ qemu_event_set(&accel_in_ioctl_event);
+}
+
+void accel_cpu_ioctl_begin(CPUState *cpu)
+{
+ if (unlikely(qemu_mutex_iothread_locked())) {
+ return;
+ }
+
+ /* block if lock is taken in kvm_ioctl_inhibit_begin() */
+ qemu_lockcnt_inc(&cpu->in_ioctl_lock);
+}
+
+void accel_cpu_ioctl_end(CPUState *cpu)
+{
+ if (unlikely(qemu_mutex_iothread_locked())) {
+ return;
+ }
+
+ qemu_lockcnt_dec(&cpu->in_ioctl_lock);
+ /* change event to SET. If event was BUSY, wake up all waiters */
+ qemu_event_set(&accel_in_ioctl_event);
+}
+
+static bool accel_has_to_wait(void)
+{
+ CPUState *cpu;
+ bool needs_to_wait = false;
+
+ CPU_FOREACH(cpu) {
+ if (qemu_lockcnt_count(&cpu->in_ioctl_lock)) {
+ /* exit the ioctl, if vcpu is running it */
+ qemu_cpu_kick(cpu);
+ needs_to_wait = true;
+ }
+ }
+
+ return needs_to_wait || qemu_lockcnt_count(&accel_in_ioctl_lock);
+}
+
+void accel_ioctl_inhibit_begin(void)
+{
+ CPUState *cpu;
+
+ /*
+ * We allow to inhibit only when holding the BQL, so we can identify
+ * when an inhibitor wants to issue an ioctl easily.
+ */
+ g_assert(qemu_mutex_iothread_locked());
+
+ /* Block further invocations of the ioctls outside the BQL. */
+ CPU_FOREACH(cpu) {
+ qemu_lockcnt_lock(&cpu->in_ioctl_lock);
+ }
+ qemu_lockcnt_lock(&accel_in_ioctl_lock);
+
+ /* Keep waiting until there are running ioctls */
+ while (true) {
+
+ /* Reset event to FREE. */
+ qemu_event_reset(&accel_in_ioctl_event);
+
+ if (accel_has_to_wait()) {
+ /*
+ * If event is still FREE, and there are ioctls still in progress,
+ * wait.
+ *
+ * If an ioctl finishes before qemu_event_wait(), it will change
+ * the event state to SET. This will prevent qemu_event_wait() from
+ * blocking, but it's not a problem because if other ioctls are
+ * still running the loop will iterate once more and reset the event
+ * status to FREE so that it can wait properly.
+ *
+ * If an ioctls finishes while qemu_event_wait() is blocking, then
+ * it will be waken up, but also here the while loop makes sure
+ * to re-enter the wait if there are other running ioctls.
+ */
+ qemu_event_wait(&accel_in_ioctl_event);
+ } else {
+ /* No ioctl is running */
+ return;
+ }
+ }
+}
+
+void accel_ioctl_inhibit_end(void)
+{
+ CPUState *cpu;
+
+ qemu_lockcnt_unlock(&accel_in_ioctl_lock);
+ CPU_FOREACH(cpu) {
+ qemu_lockcnt_unlock(&cpu->in_ioctl_lock);
+ }
+}
+
diff --git a/accel/meson.build b/accel/meson.build
index dfd808d2c8..801b4d44e8 100644
--- a/accel/meson.build
+++ b/accel/meson.build
@@ -1,4 +1,4 @@
-specific_ss.add(files('accel-common.c'))
+specific_ss.add(files('accel-common.c', 'accel-blocker.c'))
softmmu_ss.add(files('accel-softmmu.c'))
user_ss.add(files('accel-user.c'))
diff --git a/hw/core/cpu-common.c b/hw/core/cpu-common.c
index 9e3241b430..b6e83acf0a 100644
--- a/hw/core/cpu-common.c
+++ b/hw/core/cpu-common.c
@@ -238,6 +238,7 @@ static void cpu_common_initfn(Object *obj)
cpu->nr_threads = 1;
qemu_mutex_init(&cpu->work_mutex);
+ qemu_lockcnt_init(&cpu->in_ioctl_lock);
QSIMPLEQ_INIT(&cpu->work_list);
QTAILQ_INIT(&cpu->breakpoints);
QTAILQ_INIT(&cpu->watchpoints);
@@ -249,6 +250,7 @@ static void cpu_common_finalize(Object *obj)
{
CPUState *cpu = CPU(obj);
+ qemu_lockcnt_destroy(&cpu->in_ioctl_lock);
qemu_mutex_destroy(&cpu->work_mutex);
}
diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
index e948e81f1a..49d9c73f97 100644
--- a/include/hw/core/cpu.h
+++ b/include/hw/core/cpu.h
@@ -383,6 +383,9 @@ struct CPUState {
uint32_t kvm_fetch_index;
uint64_t dirty_pages;
+ /* Use by accel-block: CPU is executing an ioctl() */
+ QemuLockCnt in_ioctl_lock;
+
/* Used for events with 'vcpu' and *without* the 'disabled' properties */
DECLARE_BITMAP(trace_dstate_delayed, CPU_TRACE_DSTATE_MAX_EVENTS);
DECLARE_BITMAP(trace_dstate, CPU_TRACE_DSTATE_MAX_EVENTS);
diff --git a/include/sysemu/accel-blocker.h b/include/sysemu/accel-blocker.h
new file mode 100644
index 0000000000..72020529ef
--- /dev/null
+++ b/include/sysemu/accel-blocker.h
@@ -0,0 +1,56 @@
+/*
+ * Accelerator blocking API, to prevent new ioctls from starting and wait the
+ * running ones finish.
+ * This mechanism differs from pause/resume_all_vcpus() in that it does not
+ * release the BQL.
+ *
+ * Copyright (c) 2022 Red Hat Inc.
+ *
+ * Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+#ifndef ACCEL_BLOCKER_H
+#define ACCEL_BLOCKER_H
+
+#include "qemu/osdep.h"
+#include "sysemu/cpus.h"
+
+extern void accel_blocker_init(void);
+
+/*
+ * accel_{cpu_}ioctl_begin/end:
+ * Mark when ioctl is about to run or just finished.
+ *
+ * accel_{cpu_}ioctl_begin will block after accel_ioctl_inhibit_begin() is
+ * called, preventing new ioctls to run. They will continue only after
+ * accel_ioctl_inibith_end().
+ */
+extern void accel_ioctl_begin(void);
+extern void accel_ioctl_end(void);
+extern void accel_cpu_ioctl_begin(CPUState *cpu);
+extern void accel_cpu_ioctl_end(CPUState *cpu);
+
+/*
+ * accel_ioctl_inhibit_begin: start critical section
+ *
+ * This function makes sure that:
+ * 1) incoming accel_{cpu_}ioctl_begin() calls block
+ * 2) wait that all ioctls that were already running reach
+ * accel_{cpu_}ioctl_end(), kicking vcpus if necessary.
+ *
+ * This allows the caller to access shared data or perform operations without
+ * worrying of concurrent vcpus accesses.
+ */
+extern void accel_ioctl_inhibit_begin(void);
+
+/*
+ * accel_ioctl_inhibit_end: end critical section started by
+ * accel_ioctl_inhibit_begin()
+ *
+ * This function allows blocked accel_{cpu_}ioctl_begin() to continue.
+ */
+extern void accel_ioctl_inhibit_end(void);
+
+#endif /* ACCEL_BLOCKER_H */
diff --git a/util/meson.build b/util/meson.build
index 05b593055a..b5f153b0e8 100644
--- a/util/meson.build
+++ b/util/meson.build
@@ -48,6 +48,7 @@ util_ss.add(files('transactions.c'))
util_ss.add(when: 'CONFIG_POSIX', if_true: files('drm.c'))
util_ss.add(files('guest-random.c'))
util_ss.add(files('yank.c'))
+util_ss.add(files('lockcnt.c'))
if have_user
util_ss.add(files('selfmap.c'))
@@ -69,7 +70,6 @@ if have_block
util_ss.add(files('hexdump.c'))
util_ss.add(files('iova-tree.c'))
util_ss.add(files('iov.c', 'qemu-sockets.c', 'uri.c'))
- util_ss.add(files('lockcnt.c'))
util_ss.add(files('main-loop.c'))
util_ss.add(files('nvdimm-utils.c'))
util_ss.add(files('qemu-coroutine.c', 'qemu-coroutine-lock.c', 'qemu-coroutine-io.c'))
--
2.37.3

78
SOURCES/kvm-acpi-fix-OEM-ID-OEM-Table-ID-padding.patch
Unescape View File

@ -0,0 +1,78 @@
From af082f3499de265d123157d097b5c84981e0aa63 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Wed, 30 Mar 2022 14:52:34 -0400
Subject: [PATCH 15/18] acpi: fix OEM ID/OEM Table ID padding
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 141: acpi: fix QEMU crash when started with SLIC table
RH-Commit: [7/10] 51ea859cbe12b5a902d529ab589d18757d98f71d (jmaloy/qemu-kvm)
RH-Bugzilla: 2062611
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2062611
Upstream: Merged
commit 748c030f360a940fe0c9382c8ca1649096c3a80d
Author: Igor Mammedov <imammedo@redhat.com>
Date: Wed Jan 12 08:03:31 2022 -0500
acpi: fix OEM ID/OEM Table ID padding
Commit [2] broke original '\0' padding of OEM ID and OEM Table ID
fields in headers of ACPI tables. While it doesn't have impact on
default values since QEMU uses 6 and 8 characters long values
respectively, it broke usecase where IDs are provided on QEMU CLI.
It shouldn't affect guest (but may cause licensing verification
issues in guest OS).
One of the broken usecases is user supplied SLIC table with IDs
shorter than max possible length, where [2] mangles IDs with extra
spaces in RSDT and FADT tables whereas guest OS expects those to
mirror the respective values of the used SLIC table.
Fix it by replacing whitespace padding with '\0' padding in
accordance with [1] and expectations of guest OS
1) ACPI spec, v2.0b
17.2 AML Grammar Definition
...
//OEM ID of up to 6 characters. If the OEM ID is
//shorter than 6 characters, it can be terminated
//with a NULL character.
2)
Fixes: 602b458201 ("acpi: Permit OEM ID and OEM table ID fields to be changed")
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/707
Reported-by: Dmitry V. Orekhov <dima.orekhov@gmail.com>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Cc: qemu-stable@nongnu.org
Message-Id: <20220112130332.1648664-4-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Ani Sinha <ani@anisinha.ca>
Tested-by: Dmitry V. Orekhov dima.orekhov@gmail.com
(cherry picked from commit 748c030f360a940fe0c9382c8ca1649096c3a80d)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/acpi/aml-build.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/hw/acpi/aml-build.c b/hw/acpi/aml-build.c
index b3b3310df3..65148d5b9d 100644
--- a/hw/acpi/aml-build.c
+++ b/hw/acpi/aml-build.c
@@ -1724,9 +1724,9 @@ void acpi_table_begin(AcpiTable *desc, GArray *array)
build_append_int_noprefix(array, 0, 4); /* Length */
build_append_int_noprefix(array, desc->rev, 1); /* Revision */
build_append_int_noprefix(array, 0, 1); /* Checksum */
- build_append_padded_str(array, desc->oem_id, 6, ' '); /* OEMID */
+ build_append_padded_str(array, desc->oem_id, 6, '\0'); /* OEMID */
/* OEM Table ID */
- build_append_padded_str(array, desc->oem_table_id, 8, ' ');
+ build_append_padded_str(array, desc->oem_table_id, 8, '\0');
build_append_int_noprefix(array, 1, 4); /* OEM Revision */
g_array_append_vals(array, ACPI_BUILD_APPNAME8, 4); /* Creator ID */
build_append_int_noprefix(array, 1, 4); /* Creator Revision */
--
2.27.0

108
SOURCES/kvm-acpi-fix-QEMU-crash-when-started-with-SLIC-table.patch
Unescape View File

@ -0,0 +1,108 @@
From 4e8fb957a349558648d5cddb80a89460bc97439e Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Wed, 30 Mar 2022 14:52:34 -0400
Subject: [PATCH 09/18] acpi: fix QEMU crash when started with SLIC table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 141: acpi: fix QEMU crash when started with SLIC table
RH-Commit: [1/10] 0c34e80346c33da4f220d9c486b120c35005144e (jmaloy/qemu-kvm)
RH-Bugzilla: 2062611
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2062611
Upstream: Merged
commit 8cdb99af45365727ac17f45239a9b8c1d5155c6d)
Author: Igor Mammedov <imammedo@redhat.com>
Date: Mon Dec 27 14:31:17 2021 -0500
acpi: fix QEMU crash when started with SLIC table
if QEMU is started with used provided SLIC table blob,
-acpitable sig=SLIC,oem_id='CRASH ',oem_table_id="ME",oem_rev=00002210,asl_compiler_id="",asl_compiler_rev=00000000,data=/dev/null
it will assert with:
hw/acpi/aml-build.c:61:build_append_padded_str: assertion failed: (len <= maxlen)
and following backtrace:
...
build_append_padded_str (array=0x555556afe320, str=0x555556afdb2e "CRASH ME", maxlen=0x6, pad=0x20) at hw/acpi/aml-build.c:61
acpi_table_begin (desc=0x7fffffffd1b0, array=0x555556afe320) at hw/acpi/aml-build.c:1727
build_fadt (tbl=0x555556afe320, linker=0x555557ca3830, f=0x7fffffffd318, oem_id=0x555556afdb2e "CRASH ME", oem_table_id=0x555556afdb34 "ME") at hw/acpi/aml-build.c:2064
...
which happens due to acpi_table_begin() expecting NULL terminated
oem_id and oem_table_id strings, which is normally the case, but
in case of user provided SLIC table, oem_id points to table's blob
directly and as result oem_id became longer than expected.
Fix issue by handling oem_id consistently and make acpi_get_slic_oem()
return NULL terminated strings.
PS:
After [1] refactoring, oem_id semantics became inconsistent, where
NULL terminated string was coming from machine and old way pointer
into byte array coming from -acpitable option. That used to work
since build_header() wasn't expecting NULL terminated string and
blindly copied the 1st 6 bytes only.
However commit [2] broke that by replacing build_header() with
acpi_table_begin(), which was expecting NULL terminated string
and was checking oem_id size.
1) 602b45820 ("acpi: Permit OEM ID and OEM table ID fields to be changed")
2)
Fixes: 4b56e1e4eb08 ("acpi: build_fadt: use acpi_table_begin()/acpi_table_end() instead of build_header()")
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/786
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20211227193120.1084176-2-imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Denis Lisov <dennis.lissov@gmail.com>
Tested-by: Alexander Tsoy <alexander@tsoy.me>
Cc: qemu-stable@nongnu.org
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 8cdb99af45365727ac17f45239a9b8c1d5155c6d)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/acpi/core.c | 4 ++--
hw/i386/acpi-build.c | 2 ++
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/hw/acpi/core.c b/hw/acpi/core.c
index 1e004d0078..3e811bf03c 100644
--- a/hw/acpi/core.c
+++ b/hw/acpi/core.c
@@ -345,8 +345,8 @@ int acpi_get_slic_oem(AcpiSlicOem *oem)
struct acpi_table_header *hdr = (void *)(u - sizeof(hdr->_length));
if (memcmp(hdr->sig, "SLIC", 4) == 0) {
- oem->id = hdr->oem_id;
- oem->table_id = hdr->oem_table_id;
+ oem->id = g_strndup(hdr->oem_id, 6);
+ oem->table_id = g_strndup(hdr->oem_table_id, 8);
return 0;
}
}
diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
index a4478e77b7..acc4869db0 100644
--- a/hw/i386/acpi-build.c
+++ b/hw/i386/acpi-build.c
@@ -2726,6 +2726,8 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)
/* Cleanup memory that's no longer used. */
g_array_free(table_offsets, true);
+ g_free(slic_oem.id);
+ g_free(slic_oem.table_id);
}
static void acpi_ram_update(MemoryRegion *mr, GArray *data)
--
2.27.0

140
SOURCES/kvm-acpi-pcihp-pcie-set-power-on-cap-on-parent-slot.patch
Unescape View File

@ -0,0 +1,140 @@
From c9ceb175667cdeead59384a97a812367ae19c570 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Wed, 23 Mar 2022 13:21:40 -0400
Subject: [PATCH 06/18] acpi: pcihp: pcie: set power on cap on parent slot
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 134: pci: expose TYPE_XIO3130_DOWNSTREAM name
RH-Commit: [2/2] d883872647a6e90ec573140b2c171f3f53b600ab (jmaloy/qemu-kvm)
RH-Bugzilla: 2062610
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
BZ: https://bugzilla.redhat.com/2062610
UPSTREAM: merged
BREW: https://brewweb.engineering.redhat.com/brew/taskinfo?taskID=44038138
commit 6b0969f1ec825984cd74619f0730be421b0c46fb
Author: Igor Mammedov <imammedo@redhat.com>
Date: Tue Mar 1 10:11:59 2022 -0500
acpi: pcihp: pcie: set power on cap on parent slot
on creation a PCIDevice has power turned on at the end of pci_qdev_realize()
however later on if PCIe slot isn't populated with any children
it's power is turned off. It's fine if native hotplug is used
as plug callback will power slot on among other things.
However when ACPI hotplug is enabled it replaces native PCIe plug
callbacks with ACPI specific ones (acpi_pcihp_device_*plug_cb) and
as result slot stays powered off. It works fine as ACPI hotplug
on guest side takes care of enumerating/initializing hotplugged
device. But when later guest is migrated, call chain introduced by]
commit d5daff7d312 (pcie: implement slot power control for pcie root ports)
pcie_cap_slot_post_load()
-> pcie_cap_update_power()
-> pcie_set_power_device()
-> pci_set_power()
-> pci_update_mappings()
will disable earlier initialized BARs for the hotplugged device
in powered off slot due to commit 23786d13441 (pci: implement power state)
which disables BARs if power is off.
Fix it by setting PCI_EXP_SLTCTL_PCC to PCI_EXP_SLTCTL_PWR_ON
on slot (root port/downstream port) at the time a device
hotplugged into it. As result PCI_EXP_SLTCTL_PWR_ON is migrated
to target and above call chain keeps device plugged into it
powered on.
Fixes: d5daff7d312 ("pcie: implement slot power control for pcie root ports")
Fixes: 23786d13441 ("pci: implement power state")
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=2053584
Suggested-by: "Michael S. Tsirkin" <mst@redhat.com>
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20220301151200.3507298-3-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
(cherry picked from commit 6b0969f1ec825984cd74619f0730be421b0c46fb)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/acpi/pcihp.c | 12 +++++++++++-
hw/pci/pcie.c | 11 +++++++++++
include/hw/pci/pcie.h | 1 +
3 files changed, 23 insertions(+), 1 deletion(-)
diff --git a/hw/acpi/pcihp.c b/hw/acpi/pcihp.c
index a5e182dd3a..be0e846b34 100644
--- a/hw/acpi/pcihp.c
+++ b/hw/acpi/pcihp.c
@@ -32,6 +32,7 @@
#include "hw/pci/pci_bridge.h"
#include "hw/pci/pci_host.h"
#include "hw/pci/pcie_port.h"
+#include "hw/pci-bridge/xio3130_downstream.h"
#include "hw/i386/acpi-build.h"
#include "hw/acpi/acpi.h"
#include "hw/pci/pci_bus.h"
@@ -341,6 +342,8 @@ void acpi_pcihp_device_plug_cb(HotplugHandler *hotplug_dev, AcpiPciHpState *s,
{
PCIDevice *pdev = PCI_DEVICE(dev);
int slot = PCI_SLOT(pdev->devfn);
+ PCIDevice *bridge;
+ PCIBus *bus;
int bsel;
/* Don't send event when device is enabled during qemu machine creation:
@@ -370,7 +373,14 @@ void acpi_pcihp_device_plug_cb(HotplugHandler *hotplug_dev, AcpiPciHpState *s,
return;
}
- bsel = acpi_pcihp_get_bsel(pci_get_bus(pdev));
+ bus = pci_get_bus(pdev);
+ bridge = pci_bridge_get_device(bus);
+ if (object_dynamic_cast(OBJECT(bridge), TYPE_PCIE_ROOT_PORT) ||
+ object_dynamic_cast(OBJECT(bridge), TYPE_XIO3130_DOWNSTREAM)) {
+ pcie_cap_slot_enable_power(bridge);
+ }
+
+ bsel = acpi_pcihp_get_bsel(bus);
g_assert(bsel >= 0);
s->acpi_pcihp_pci_status[bsel].up |= (1U << slot);
acpi_send_event(DEVICE(hotplug_dev), ACPI_PCI_HOTPLUG_STATUS);
diff --git a/hw/pci/pcie.c b/hw/pci/pcie.c
index d7d73a31e4..996f0e24fe 100644
--- a/hw/pci/pcie.c
+++ b/hw/pci/pcie.c
@@ -366,6 +366,17 @@ static void hotplug_event_clear(PCIDevice *dev)
}
}
+void pcie_cap_slot_enable_power(PCIDevice *dev)
+{
+ uint8_t *exp_cap = dev->config + dev->exp.exp_cap;
+ uint32_t sltcap = pci_get_long(exp_cap + PCI_EXP_SLTCAP);
+
+ if (sltcap & PCI_EXP_SLTCAP_PCP) {
+ pci_set_word_by_mask(exp_cap + PCI_EXP_SLTCTL,
+ PCI_EXP_SLTCTL_PCC, PCI_EXP_SLTCTL_PWR_ON);
+ }
+}
+
static void pcie_set_power_device(PCIBus *bus, PCIDevice *dev, void *opaque)
{
bool *power = opaque;
diff --git a/include/hw/pci/pcie.h b/include/hw/pci/pcie.h
index 6063bee0ec..c27368d077 100644
--- a/include/hw/pci/pcie.h
+++ b/include/hw/pci/pcie.h
@@ -112,6 +112,7 @@ void pcie_cap_slot_write_config(PCIDevice *dev,
uint32_t addr, uint32_t val, int len);
int pcie_cap_slot_post_load(void *opaque, int version_id);
void pcie_cap_slot_push_attention_button(PCIDevice *dev);
+void pcie_cap_slot_enable_power(PCIDevice *dev);
void pcie_cap_root_init(PCIDevice *dev);
void pcie_cap_root_reset(PCIDevice *dev);
--
2.27.0

51
SOURCES/kvm-acpi-validate-hotplug-selector-on-access.patch
Unescape View File

@ -0,0 +1,51 @@
From 529a5d908f5d16714b8ae0a51eaaaa84994dfae8 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Tue, 21 Dec 2021 09:45:44 -0500
Subject: [PATCH 1/2] acpi: validate hotplug selector on access
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 97: acpi: validate hotplug selector on access
RH-Commit: [1/1] 79bcfb0df0091e2b716d2e1c545f047b3409c26c (jmaloy/qemu-kvm)
RH-Bugzilla: 2036580
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Daniel P. Berrangé <berrange@redhat.com>
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
When bus is looked up on a pci write, we didn't
validate that the lookup succeeded.
Fuzzers thus can trigger QEMU crash by dereferencing the NULL
bus pointer.
Fixes: b32bd763a1 ("pci: introduce acpi-index property for PCI device")
Fixes: CVE-2021-4158
Cc: "Igor Mammedov" <imammedo@redhat.com>
Fixes: https://gitlab.com/qemu-project/qemu/-/issues/770
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Ani Sinha <ani@anisinha.ca>
(cherry picked from commit 9bd6565ccee68f72d5012e24646e12a1c662827e)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/acpi/pcihp.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/hw/acpi/pcihp.c b/hw/acpi/pcihp.c
index 30405b5113..a5e182dd3a 100644
--- a/hw/acpi/pcihp.c
+++ b/hw/acpi/pcihp.c
@@ -491,6 +491,9 @@ static void pci_write(void *opaque, hwaddr addr, uint64_t data,
}
bus = acpi_pcihp_find_hotplug_bus(s, s->hotplug_select);
+ if (!bus) {
+ break;
+ }
QTAILQ_FOREACH_SAFE(kid, &bus->qbus.children, sibling, next) {
Object *o = OBJECT(kid->child);
PCIDevice *dev = PCI_DEVICE(o);
--
2.27.0

50
SOURCES/kvm-aio-wait-switch-to-smp_mb__after_rmw.patch
Unescape View File

@ -0,0 +1,50 @@
From 953c5c0982b61b0a3f8f03452844b5487eb22fc7 Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Thu, 9 Mar 2023 08:13:17 -0500
Subject: [PATCH 06/13] aio-wait: switch to smp_mb__after_rmw()
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 263: qatomic: add smp_mb__before/after_rmw()
RH-Bugzilla: 2168472
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Commit: [6/10] 9f30f97754139ffd18d36b2350f9ed4e59ac496e
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2168472
commit b532526a07ef3b903ead2e055fe6cc87b41057a3
Author: Paolo Bonzini <pbonzini@redhat.com>
Date: Fri Mar 3 11:03:52 2023 +0100
aio-wait: switch to smp_mb__after_rmw()
The barrier comes after an atomic increment, so it is enough to use
smp_mb__after_rmw(); this avoids a double barrier on x86 systems.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
include/block/aio-wait.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/block/aio-wait.h b/include/block/aio-wait.h
index 54840f8622..03b6394c78 100644
--- a/include/block/aio-wait.h
+++ b/include/block/aio-wait.h
@@ -82,7 +82,7 @@ extern AioWait global_aio_wait;
/* Increment wait_->num_waiters before evaluating cond. */ \
qatomic_inc(&wait_->num_waiters); \
/* Paired with smp_mb in aio_wait_kick(). */ \
- smp_mb(); \
+ smp_mb__after_rmw(); \
if (ctx_ && in_aio_context_home_thread(ctx_)) { \
while ((cond)) { \
aio_poll(ctx_, true); \
--
2.37.3

86
SOURCES/kvm-aio_wait_kick-add-missing-memory-barrier.patch
Unescape View File

@ -0,0 +1,86 @@
From d7eae0ff4c7f7f7bf10f10272adf7c6971c0db9b Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Thu, 9 Mar 2023 09:26:35 -0500
Subject: [PATCH 01/13] aio_wait_kick: add missing memory barrier
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 263: qatomic: add smp_mb__before/after_rmw()
RH-Bugzilla: 2168472
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Commit: [1/10] eb774aee79864052e14e706d931e52e7bd1162c8
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2168472
commit 7455ff1aa01564cc175db5b2373e610503ad4411
Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Tue May 24 13:30:54 2022 -0400
aio_wait_kick: add missing memory barrier
It seems that aio_wait_kick always required a memory barrier
or atomic operation in the caller, but nobody actually
took care of doing it.
Let's put the barrier in the function instead, and pair it
with another one in AIO_WAIT_WHILE. Read aio_wait_kick()
comment for further explanation.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Message-Id: <20220524173054.12651-1-eesposit@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
include/block/aio-wait.h | 2 ++
util/aio-wait.c | 16 +++++++++++++++-
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/include/block/aio-wait.h b/include/block/aio-wait.h
index b39eefb38d..54840f8622 100644
--- a/include/block/aio-wait.h
+++ b/include/block/aio-wait.h
@@ -81,6 +81,8 @@ extern AioWait global_aio_wait;
AioContext *ctx_ = (ctx); \
/* Increment wait_->num_waiters before evaluating cond. */ \
qatomic_inc(&wait_->num_waiters); \
+ /* Paired with smp_mb in aio_wait_kick(). */ \
+ smp_mb(); \
if (ctx_ && in_aio_context_home_thread(ctx_)) { \
while ((cond)) { \
aio_poll(ctx_, true); \
diff --git a/util/aio-wait.c b/util/aio-wait.c
index bdb3d3af22..98c5accd29 100644
--- a/util/aio-wait.c
+++ b/util/aio-wait.c
@@ -35,7 +35,21 @@ static void dummy_bh_cb(void *opaque)
void aio_wait_kick(void)
{
- /* The barrier (or an atomic op) is in the caller. */
+ /*
+ * Paired with smp_mb in AIO_WAIT_WHILE. Here we have:
+ * write(condition);
+ * aio_wait_kick() {
+ * smp_mb();
+ * read(num_waiters);
+ * }
+ *
+ * And in AIO_WAIT_WHILE:
+ * write(num_waiters);
+ * smp_mb();
+ * read(condition);
+ */
+ smp_mb();
+
if (qatomic_read(&global_aio_wait.num_waiters)) {
aio_bh_schedule_oneshot(qemu_get_aio_context(), dummy_bh_cb, NULL);
}
--
2.37.3

66
SOURCES/kvm-async-clarify-usage-of-barriers-in-the-polling-case.patch
Unescape View File

@ -0,0 +1,66 @@
From 187eb7a418af93375e42298d06e231e2bec3cf00 Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Thu, 9 Mar 2023 08:15:42 -0500
Subject: [PATCH 10/13] async: clarify usage of barriers in the polling case
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 263: qatomic: add smp_mb__before/after_rmw()
RH-Bugzilla: 2168472
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Commit: [10/10] 3be07ccc6137a0336becfe63a818d9cbadb38e9c
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2168472
commit 6229438cca037d42f44a96d38feb15cb102a444f
Author: Paolo Bonzini <pbonzini@redhat.com>
Date: Mon Mar 6 10:43:52 2023 +0100
async: clarify usage of barriers in the polling case
Explain that aio_context_notifier_poll() relies on
aio_notify_accept() to catch all the memory writes that were
done before ctx->notified was set to true.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
util/async.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/util/async.c b/util/async.c
index 795fe699b6..2a63bf90f2 100644
--- a/util/async.c
+++ b/util/async.c
@@ -463,8 +463,9 @@ void aio_notify_accept(AioContext *ctx)
qatomic_set(&ctx->notified, false);
/*
- * Write ctx->notified before reading e.g. bh->flags. Pairs with smp_wmb
- * in aio_notify.
+ * Order reads of ctx->notified (in aio_context_notifier_poll()) and the
+ * above clearing of ctx->notified before reads of e.g. bh->flags. Pairs
+ * with smp_wmb() in aio_notify.
*/
smp_mb();
}
@@ -487,6 +488,11 @@ static bool aio_context_notifier_poll(void *opaque)
EventNotifier *e = opaque;
AioContext *ctx = container_of(e, AioContext, notifier);
+ /*
+ * No need for load-acquire because we just want to kick the
+ * event loop. aio_notify_accept() takes care of synchronizing
+ * the event loop with the producers.
+ */
return qatomic_read(&ctx->notified);
}
--
2.37.3

111
SOURCES/kvm-async-update-documentation-of-the-memory-barriers.patch
Unescape View File

@ -0,0 +1,111 @@
From ea3856bb545d19499602830cdc3076d83a981e7a Mon Sep 17 00:00:00 2001
From: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Date: Thu, 9 Mar 2023 08:15:36 -0500
Subject: [PATCH 09/13] async: update documentation of the memory barriers
RH-Author: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-MergeRequest: 263: qatomic: add smp_mb__before/after_rmw()
RH-Bugzilla: 2168472
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Paolo Bonzini <pbonzini@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Commit: [9/10] d471da2acf7a107cf75f3327c5e8d7456307160e
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2168472
commit 8dd48650b43dfde4ebea34191ac267e474bcc29e
Author: Paolo Bonzini <pbonzini@redhat.com>
Date: Mon Mar 6 10:15:06 2023 +0100
async: update documentation of the memory barriers
Ever since commit 8c6b0356b539 ("util/async: make bh_aio_poll() O(1)",
2020-02-22), synchronization between qemu_bh_schedule() and aio_bh_poll()
is happening when the bottom half is enqueued in the bh_list; not
when the flags are set. Update the documentation to match.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
util/async.c | 33 +++++++++++++++++++--------------
1 file changed, 19 insertions(+), 14 deletions(-)
diff --git a/util/async.c b/util/async.c
index 6f6717a34b..795fe699b6 100644
--- a/util/async.c
+++ b/util/async.c
@@ -71,14 +71,21 @@ static void aio_bh_enqueue(QEMUBH *bh, unsigned new_flags)
unsigned old_flags;
/*
- * The memory barrier implicit in qatomic_fetch_or makes sure that:
- * 1. idle & any writes needed by the callback are done before the
- * locations are read in the aio_bh_poll.
- * 2. ctx is loaded before the callback has a chance to execute and bh
- * could be freed.
+ * Synchronizes with atomic_fetch_and() in aio_bh_dequeue(), ensuring that
+ * insertion starts after BH_PENDING is set.
*/
old_flags = qatomic_fetch_or(&bh->flags, BH_PENDING | new_flags);
+
if (!(old_flags & BH_PENDING)) {
+ /*
+ * At this point the bottom half becomes visible to aio_bh_poll().
+ * This insertion thus synchronizes with QSLIST_MOVE_ATOMIC in
+ * aio_bh_poll(), ensuring that:
+ * 1. any writes needed by the callback are visible from the callback
+ * after aio_bh_dequeue() returns bh.
+ * 2. ctx is loaded before the callback has a chance to execute and bh
+ * could be freed.
+ */
QSLIST_INSERT_HEAD_ATOMIC(&ctx->bh_list, bh, next);
}
@@ -97,11 +104,8 @@ static QEMUBH *aio_bh_dequeue(BHList *head, unsigned *flags)
QSLIST_REMOVE_HEAD(head, next);
/*
- * The qatomic_and is paired with aio_bh_enqueue(). The implicit memory
- * barrier ensures that the callback sees all writes done by the scheduling
- * thread. It also ensures that the scheduling thread sees the cleared
- * flag before bh->cb has run, and thus will call aio_notify again if
- * necessary.
+ * Synchronizes with qatomic_fetch_or() in aio_bh_enqueue(), ensuring that
+ * the removal finishes before BH_PENDING is reset.
*/
*flags = qatomic_fetch_and(&bh->flags,
~(BH_PENDING | BH_SCHEDULED | BH_IDLE));
@@ -148,6 +152,7 @@ int aio_bh_poll(AioContext *ctx)
BHListSlice *s;
int ret = 0;
+ /* Synchronizes with QSLIST_INSERT_HEAD_ATOMIC in aio_bh_enqueue(). */
QSLIST_MOVE_ATOMIC(&slice.bh_list, &ctx->bh_list);
QSIMPLEQ_INSERT_TAIL(&ctx->bh_slice_list, &slice, next);
@@ -437,15 +442,15 @@ LuringState *aio_get_linux_io_uring(AioContext *ctx)
void aio_notify(AioContext *ctx)
{
/*
- * Write e.g. bh->flags before writing ctx->notified. Pairs with smp_mb in
- * aio_notify_accept.
+ * Write e.g. ctx->bh_list before writing ctx->notified. Pairs with
+ * smp_mb() in aio_notify_accept().
*/
smp_wmb();
qatomic_set(&ctx->notified, true);
/*
- * Write ctx->notified before reading ctx->notify_me. Pairs
- * with smp_mb in aio_ctx_prepare or aio_poll.
+ * Write ctx->notified (and also ctx->bh_list) before reading ctx->notify_me.
+ * Pairs with smp_mb() in aio_ctx_prepare or aio_poll.
*/
smp_mb();
if (qatomic_read(&ctx->notify_me)) {
--
2.37.3

71
SOURCES/kvm-backends-hostmem-Fix-support-of-memory-backend-memfd.patch
Unescape View File

@ -0,0 +1,71 @@
From 60da56e3685969493ae483c3cc2c66af13d00baf Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Wed, 10 Aug 2022 14:57:18 +0200
Subject: [PATCH 1/3] backends/hostmem: Fix support of memory-backend-memfd in
qemu_maxrampagesize()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <None>
RH-MergeRequest: 221: backends/hostmem: Fix support of memory-backend-memfd in qemu_maxrampagesize()
RH-Bugzilla: 2117149
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Commit: [1/1] b5a1047750af32c0a261b8385ea0e819eb16681a
It is currently not possible yet to use "memory-backend-memfd" on s390x
with hugepages enabled. This problem is caused by qemu_maxrampagesize()
not taking memory-backend-memfd objects into account yet, so the code
in s390_memory_init() fails to enable the huge page support there via
s390_set_max_pagesize(). Fix it by generalizing the code, so that it
looks at qemu_ram_pagesize(memdev->mr.ram_block) instead of re-trying
to get the information from the filesystem.
Suggested-by: David Hildenbrand <david@redhat.com>
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2116496
Message-Id: <20220810125720.3849835-2-thuth@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 8be934b70e923104da883b990dee18f02552d40e)
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2117149
[clg: Resolved conflict on qemu_real_host_page_size() ]
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
backends/hostmem.c | 14 ++------------
1 file changed, 2 insertions(+), 12 deletions(-)
diff --git a/backends/hostmem.c b/backends/hostmem.c
index 4c05862ed5..0c4654ea85 100644
--- a/backends/hostmem.c
+++ b/backends/hostmem.c
@@ -305,22 +305,12 @@ bool host_memory_backend_is_mapped(HostMemoryBackend *backend)
return backend->is_mapped;
}
-#ifdef __linux__
size_t host_memory_backend_pagesize(HostMemoryBackend *memdev)
{
- Object *obj = OBJECT(memdev);
- char *path = object_property_get_str(obj, "mem-path", NULL);
- size_t pagesize = qemu_mempath_getpagesize(path);
-
- g_free(path);
+ size_t pagesize = qemu_ram_pagesize(memdev->mr.ram_block);
+ g_assert(pagesize >= qemu_real_host_page_size);
return pagesize;
}
-#else
-size_t host_memory_backend_pagesize(HostMemoryBackend *memdev)
-{
- return qemu_real_host_page_size;
-}
-#endif
static void
host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
--
2.35.3

63
SOURCES/kvm-block-Lock-AioContext-for-drain_end-in-blockdev-reop.patch
Unescape View File

@ -0,0 +1,63 @@
From b21fa5ecd9acf2b91839a2915fb4bb39dac4c803 Mon Sep 17 00:00:00 2001
From: Kevin Wolf <kwolf@redhat.com>
Date: Thu, 3 Feb 2022 15:05:33 +0100
Subject: [PATCH 2/5] block: Lock AioContext for drain_end in blockdev-reopen
RH-Author: Kevin Wolf <kwolf@redhat.com>
RH-MergeRequest: 142: block: Lock AioContext for drain_end in blockdev-reopen
RH-Commit: [1/2] 98de3b5987f88ea6b4b503f623d6c4475574e037
RH-Bugzilla: 2067118
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Hanna Reitz <hreitz@redhat.com>
bdrv_subtree_drained_end() requires the caller to hold the AioContext
lock for the drained node. Not doing this for nodes outside of the main
AioContext leads to crashes when AIO_WAIT_WHILE() needs to wait and
tries to temporarily release the lock.
Fixes: 3908b7a8994fa5ef7a89aa58cd5a02fc58141592
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2046659
Reported-by: Qing Wang <qinwang@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20220203140534.36522-2-kwolf@redhat.com>
Reviewed-by: Hanna Reitz <hreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit aba8205be0707b9d108e32254e186ba88107a869)
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
---
blockdev.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/blockdev.c b/blockdev.c
index b35072644e..565f6a81fd 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -3562,6 +3562,7 @@ void qmp_blockdev_reopen(BlockdevOptionsList *reopen_list, Error **errp)
{
BlockReopenQueue *queue = NULL;
GSList *drained = NULL;
+ GSList *p;
/* Add each one of the BDS that we want to reopen to the queue */
for (; reopen_list != NULL; reopen_list = reopen_list->next) {
@@ -3611,7 +3612,15 @@ void qmp_blockdev_reopen(BlockdevOptionsList *reopen_list, Error **errp)
fail:
bdrv_reopen_queue_free(queue);
- g_slist_free_full(drained, (GDestroyNotify) bdrv_subtree_drained_end);
+ for (p = drained; p; p = p->next) {
+ BlockDriverState *bs = p->data;
+ AioContext *ctx = bdrv_get_aio_context(bs);
+
+ aio_context_acquire(ctx);
+ bdrv_subtree_drained_end(bs);
+ aio_context_release(ctx);
+ }
+ g_slist_free(drained);
}
void qmp_blockdev_del(const char *node_name, Error **errp)
--
2.27.0

78
SOURCES/kvm-block-Make-bdrv_refresh_limits-non-recursive.patch
Unescape View File

@ -0,0 +1,78 @@
From 6348063b91b2370cc27153fd58fd11a6681631f6 Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Wed, 16 Feb 2022 11:53:53 +0100
Subject: [PATCH 22/24] block: Make bdrv_refresh_limits() non-recursive
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Hanna Reitz <hreitz@redhat.com>
RH-MergeRequest: 189: block: Make bdrv_refresh_limits() non-recursive
RH-Commit: [1/3] 1a1fe37f8d8f0344dd8639d6cc9d884d1aff9096
RH-Bugzilla: 2072932
RH-Acked-by: Eric Blake <eblake@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
bdrv_refresh_limits() recurses down to the node's children. That does
not seem necessary: When we refresh limits on some node, and then
recurse down and were to change one of its children's BlockLimits, then
that would mean we noticed the changed limits by pure chance. The fact
that we refresh the parent's limits has nothing to do with it, so the
reason for the change probably happened before this point in time, and
we should have refreshed the limits then.
Consequently, we should actually propagate block limits changes upwards,
not downwards.  That is a separate and pre-existing issue, though, and
so will not be addressed in this patch.
The problem with recursing is that bdrv_refresh_limits() is not atomic.
It begins with zeroing BDS.bl, and only then sets proper, valid limits.
If we do not drain all nodes whose limits are refreshed, then concurrent
I/O requests can encounter invalid request_alignment values and crash
qemu. Therefore, a recursing bdrv_refresh_limits() requires the whole
subtree to be drained, which is currently not ensured by most callers.
A non-recursive bdrv_refresh_limits() only requires the node in question
to not receive I/O requests, and this is done by most callers in some
way or another:
- bdrv_open_driver() deals with a new node with no parents yet
- bdrv_set_file_or_backing_noperm() acts on a drained node
- bdrv_reopen_commit() acts only on drained nodes
- bdrv_append() should in theory require the node to be drained; in
practice most callers just lock the AioContext, which should at least
be enough to prevent concurrent I/O requests from accessing invalid
limits
So we can resolve the bug by making bdrv_refresh_limits() non-recursive.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1879437
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20220216105355.30729-2-hreitz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 4d378bbd831bdd2f6e6adcd4ea5b77b6effaa627)
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
---
block/io.c | 4 ----
1 file changed, 4 deletions(-)
diff --git a/block/io.c b/block/io.c
index 4e4cb556c5..c3e7301613 100644
--- a/block/io.c
+++ b/block/io.c
@@ -189,10 +189,6 @@ void bdrv_refresh_limits(BlockDriverState *bs, Transaction *tran, Error **errp)
QLIST_FOREACH(c, &bs->children, next) {
if (c->role & (BDRV_CHILD_DATA | BDRV_CHILD_FILTERED | BDRV_CHILD_COW))
{
- bdrv_refresh_limits(c->bs, tran, errp);
- if (*errp) {
- return;
- }
bdrv_merge_limits(&bs->bl, &c->bs->bl);
have_limits = true;
}
--
2.35.3

129
SOURCES/kvm-block-backend-prevent-dangling-BDS-pointers-across-a.patch
Unescape View File

@ -0,0 +1,129 @@
From bf4c15a3debbe68b6eb25c52174843470a9c014f Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Tue, 11 Jan 2022 15:36:12 +0000
Subject: [PATCH 3/6] block-backend: prevent dangling BDS pointers across
aio_poll()
RH-Author: Stefan Hajnoczi <stefanha@redhat.com>
RH-MergeRequest: 109: block-backend: prevent dangling BDS pointers across aio_poll()
RH-Commit: [1/2] da5a59eddff0dc10be7de8e291fa675143d11d73
RH-Bugzilla: 2021778 2036178
RH-Acked-by: Hanna Reitz <hreitz@redhat.com>
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
The BlockBackend root child can change when aio_poll() is invoked. This
happens when a temporary filter node is removed upon blockjob
completion, for example.
Functions in block/block-backend.c must be aware of this when using a
blk_bs() pointer across aio_poll() because the BlockDriverState refcnt
may reach 0, resulting in a stale pointer.
One example is scsi_device_purge_requests(), which calls blk_drain() to
wait for in-flight requests to cancel. If the backup blockjob is active,
then the BlockBackend root child is a temporary filter BDS owned by the
blockjob. The blockjob can complete during bdrv_drained_begin() and the
last reference to the BDS is released when the temporary filter node is
removed. This results in a use-after-free when blk_drain() calls
bdrv_drained_end(bs) on the dangling pointer.
Explicitly hold a reference to bs across block APIs that invoke
aio_poll().
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2021778
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2036178
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20220111153613.25453-2-stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 1e3552dbd28359d35967b7c28dc86cde1bc29205)
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
block/block-backend.c | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/block/block-backend.c b/block/block-backend.c
index 12ef80ea17..23e727199b 100644
--- a/block/block-backend.c
+++ b/block/block-backend.c
@@ -822,16 +822,22 @@ BlockBackend *blk_by_public(BlockBackendPublic *public)
void blk_remove_bs(BlockBackend *blk)
{
ThrottleGroupMember *tgm = &blk->public.throttle_group_member;
- BlockDriverState *bs;
BdrvChild *root;
notifier_list_notify(&blk->remove_bs_notifiers, blk);
if (tgm->throttle_state) {
- bs = blk_bs(blk);
+ BlockDriverState *bs = blk_bs(blk);
+
+ /*
+ * Take a ref in case blk_bs() changes across bdrv_drained_begin(), for
+ * example, if a temporary filter node is removed by a blockjob.
+ */
+ bdrv_ref(bs);
bdrv_drained_begin(bs);
throttle_group_detach_aio_context(tgm);
throttle_group_attach_aio_context(tgm, qemu_get_aio_context());
bdrv_drained_end(bs);
+ bdrv_unref(bs);
}
blk_update_root_state(blk);
@@ -1705,6 +1711,7 @@ void blk_drain(BlockBackend *blk)
BlockDriverState *bs = blk_bs(blk);
if (bs) {
+ bdrv_ref(bs);
bdrv_drained_begin(bs);
}
@@ -1714,6 +1721,7 @@ void blk_drain(BlockBackend *blk)
if (bs) {
bdrv_drained_end(bs);
+ bdrv_unref(bs);
}
}
@@ -2044,10 +2052,13 @@ static int blk_do_set_aio_context(BlockBackend *blk, AioContext *new_context,
int ret;
if (bs) {
+ bdrv_ref(bs);
+
if (update_root_node) {
ret = bdrv_child_try_set_aio_context(bs, new_context, blk->root,
errp);
if (ret < 0) {
+ bdrv_unref(bs);
return ret;
}
}
@@ -2057,6 +2068,8 @@ static int blk_do_set_aio_context(BlockBackend *blk, AioContext *new_context,
throttle_group_attach_aio_context(tgm, new_context);
bdrv_drained_end(bs);
}
+
+ bdrv_unref(bs);
}
blk->ctx = new_context;
@@ -2326,11 +2339,13 @@ void blk_io_limits_disable(BlockBackend *blk)
ThrottleGroupMember *tgm = &blk->public.throttle_group_member;
assert(tgm->throttle_state);
if (bs) {
+ bdrv_ref(bs);
bdrv_drained_begin(bs);
}
throttle_group_unregister_tgm(tgm);
if (bs) {
bdrv_drained_end(bs);
+ bdrv_unref(bs);
}
}
--
2.27.0

56
SOURCES/kvm-block-io-Update-BSC-only-if-want_zero-is-true.patch
Unescape View File

@ -0,0 +1,56 @@
From 4c6eff78f4b31ec4bd7b42440396760d19fde63e Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Tue, 18 Jan 2022 17:59:59 +0100
Subject: [PATCH 6/7] block/io: Update BSC only if want_zero is true
RH-Author: Hanna Reitz <hreitz@redhat.com>
RH-MergeRequest: 112: block/io: Update BSC only if want_zero is true
RH-Commit: [1/2] a202de1f52110d1e871c3b5b58f2d9e9b5d17570
RH-Bugzilla: 2041480
RH-Acked-by: Eric Blake <eblake@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
We update the block-status cache whenever we get new information from a
bdrv_co_block_status() call to the block driver. However, if we have
passed want_zero=false to that call, it may flag areas containing zeroes
as data, and so we would update the block-status cache with wrong
information.
Therefore, we should not update the cache with want_zero=false.
Reported-by: Nir Soffer <nsoffer@redhat.com>
Fixes: 0bc329fbb00 ("block: block-status cache for data regions")
Reviewed-by: Nir Soffer <nsoffer@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Message-Id: <20220118170000.49423-2-hreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
(cherry picked from commit 113b727ce788335cf76f65355d670c9bc130fd75)
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
---
block/io.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/block/io.c b/block/io.c
index bb0a254def..4e4cb556c5 100644
--- a/block/io.c
+++ b/block/io.c
@@ -2497,8 +2497,12 @@ static int coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
* non-protocol nodes, and then it is never used. However, filling
* the cache requires an RCU update, so double check here to avoid
* such an update if possible.
+ *
+ * Check want_zero, because we only want to update the cache when we
+ * have accurate information about what is zero and what is data.
*/
- if (ret == (BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID) &&
+ if (want_zero &&
+ ret == (BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID) &&
QLIST_EMPTY(&bs->children))
{
/*
--
2.27.0

153
SOURCES/kvm-block-mirror-Do-not-wait-for-active-writes.patch
Unescape View File

@ -0,0 +1,153 @@
From 192f956f2b0761f270070555f8feb1f0544e5558 Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Wed, 9 Nov 2022 17:54:48 +0100
Subject: [PATCH 01/11] block/mirror: Do not wait for active writes
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 246: block/mirror: Make active mirror progress even under full load
RH-Bugzilla: 2125119
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Stefano Garzarella <sgarzare@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Commit: [1/3] 652d1e55b954f13eaec2c86f58735d4942837e16
Waiting for all active writes to settle before daring to create a
background copying operation means that we will never do background
operations while the guest does anything (in write-blocking mode), and
therefore cannot converge. Yes, we also will not diverge, but actually
converging would be even nicer.
It is unclear why we did decide to wait for all active writes to settle
before creating a background operation, but it just does not seem
necessary. Active writes will put themselves into the in_flight bitmap
and thus properly block actually conflicting background requests.
It is important for active requests to wait on overlapping background
requests, which we do in active_write_prepare(). However, so far it was
not documented why it is important. Add such documentation now, and
also to the other call of mirror_wait_on_conflicts(), so that it becomes
more clear why and when requests need to actively wait for other
requests to settle.
Another thing to note is that of course we need to ensure that there are
no active requests when the job completes, but that is done by virtue of
the BDS being drained anyway, so there cannot be any active requests at
that point.
With this change, we will need to explicitly keep track of how many
bytes are in flight in active requests so that
job_progress_set_remaining() in mirror_run() can set the correct number
of remaining bytes.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2123297
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Message-Id: <20221109165452.67927-2-hreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit d69a879bdf1aed586478eaa161ee064fe1b92f1a)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/mirror.c | 37 ++++++++++++++++++++++++++++++-------
1 file changed, 30 insertions(+), 7 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index efec2c7674..282f428cb7 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -81,6 +81,7 @@ typedef struct MirrorBlockJob {
int max_iov;
bool initial_zeroing_ongoing;
int in_active_write_counter;
+ int64_t active_write_bytes_in_flight;
bool prepared;
bool in_drain;
} MirrorBlockJob;
@@ -493,6 +494,13 @@ static uint64_t coroutine_fn mirror_iteration(MirrorBlockJob *s)
}
bdrv_dirty_bitmap_unlock(s->dirty_bitmap);
+ /*
+ * Wait for concurrent requests to @offset. The next loop will limit the
+ * copied area based on in_flight_bitmap so we only copy an area that does
+ * not overlap with concurrent in-flight requests. Still, we would like to
+ * copy something, so wait until there are at least no more requests to the
+ * very beginning of the area.
+ */
mirror_wait_on_conflicts(NULL, s, offset, 1);
job_pause_point(&s->common.job);
@@ -993,12 +1001,6 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
int64_t cnt, delta;
bool should_complete;
- /* Do not start passive operations while there are active
- * writes in progress */
- while (s->in_active_write_counter) {
- mirror_wait_for_any_operation(s, true);
- }
-
if (s->ret < 0) {
ret = s->ret;
goto immediate_exit;
@@ -1015,7 +1017,9 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
/* cnt is the number of dirty bytes remaining and s->bytes_in_flight is
* the number of bytes currently being processed; together those are
* the current remaining operation length */
- job_progress_set_remaining(&s->common.job, s->bytes_in_flight + cnt);
+ job_progress_set_remaining(&s->common.job,
+ s->bytes_in_flight + cnt +
+ s->active_write_bytes_in_flight);
/* Note that even when no rate limit is applied we need to yield
* periodically with no pending I/O so that bdrv_drain_all() returns.
@@ -1073,6 +1077,10 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
s->in_drain = true;
bdrv_drained_begin(bs);
+
+ /* Must be zero because we are drained */
+ assert(s->in_active_write_counter == 0);
+
cnt = bdrv_get_dirty_count(s->dirty_bitmap);
if (cnt > 0 || mirror_flush(s) < 0) {
bdrv_drained_end(bs);
@@ -1306,6 +1314,7 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
}
job_progress_increase_remaining(&job->common.job, bytes);
+ job->active_write_bytes_in_flight += bytes;
switch (method) {
case MIRROR_METHOD_COPY:
@@ -1327,6 +1336,7 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
abort();
}
+ job->active_write_bytes_in_flight -= bytes;
if (ret >= 0) {
job_progress_update(&job->common.job, bytes);
} else {
@@ -1375,6 +1385,19 @@ static MirrorOp *coroutine_fn active_write_prepare(MirrorBlockJob *s,
s->in_active_write_counter++;
+ /*
+ * Wait for concurrent requests affecting the area. If there are already
+ * running requests that are copying off now-to-be stale data in the area,
+ * we must wait for them to finish before we begin writing fresh data to the
+ * target so that the write operations appear in the correct order.
+ * Note that background requests (see mirror_iteration()) in contrast only
+ * wait for conflicting requests at the start of the dirty area, and then
+ * (based on the in_flight_bitmap) truncate the area to copy so it will not
+ * conflict with any requests beyond that. For active writes, however, we
+ * cannot truncate that area. The request from our parent must be blocked
+ * until the area is copied in full. Therefore, we must wait for the whole
+ * area to become free of concurrent requests.
+ */
mirror_wait_on_conflicts(op, s, offset, bytes);
bitmap_set(s->in_flight_bitmap, start_chunk, end_chunk - start_chunk);
--
2.37.3

76
SOURCES/kvm-block-mirror-Drop-mirror_wait_for_any_operation.patch
Unescape View File

@ -0,0 +1,76 @@
From 57c79ed20cb73aa9aa4dd7487379b85ea3f936f6 Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Wed, 9 Nov 2022 17:54:49 +0100
Subject: [PATCH 02/11] block/mirror: Drop mirror_wait_for_any_operation()
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 246: block/mirror: Make active mirror progress even under full load
RH-Bugzilla: 2125119
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Stefano Garzarella <sgarzare@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Commit: [2/3] dec37883bcc491441ae08d9592d1ec26a47765c0
mirror_wait_for_free_in_flight_slot() is the only remaining user of
mirror_wait_for_any_operation(), so inline the latter into the former.
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Message-Id: <20221109165452.67927-3-hreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit eb994912993077f178ccb43b20e422ecf9ae4ac7)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/mirror.c | 21 ++++++++-------------
1 file changed, 8 insertions(+), 13 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 282f428cb7..6b02555ad7 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -304,19 +304,21 @@ static int mirror_cow_align(MirrorBlockJob *s, int64_t *offset,
}
static inline void coroutine_fn
-mirror_wait_for_any_operation(MirrorBlockJob *s, bool active)
+mirror_wait_for_free_in_flight_slot(MirrorBlockJob *s)
{
MirrorOp *op;
QTAILQ_FOREACH(op, &s->ops_in_flight, next) {
- /* Do not wait on pseudo ops, because it may in turn wait on
+ /*
+ * Do not wait on pseudo ops, because it may in turn wait on
* some other operation to start, which may in fact be the
* caller of this function. Since there is only one pseudo op
* at any given time, we will always find some real operation
- * to wait on. */
- if (!op->is_pseudo_op && op->is_in_flight &&
- op->is_active_write == active)
- {
+ * to wait on.
+ * Also, do not wait on active operations, because they do not
+ * use up in-flight slots.
+ */
+ if (!op->is_pseudo_op && op->is_in_flight && !op->is_active_write) {
qemu_co_queue_wait(&op->waiting_requests, NULL);
return;
}
@@ -324,13 +326,6 @@ mirror_wait_for_any_operation(MirrorBlockJob *s, bool active)
abort();
}
-static inline void coroutine_fn
-mirror_wait_for_free_in_flight_slot(MirrorBlockJob *s)
-{
- /* Only non-active operations use up in-flight slots */
- mirror_wait_for_any_operation(s, false);
-}
-
/* Perform a mirror copy operation.
*
* *op->bytes_handled is set to the number of bytes copied after and
--
2.37.3

75
SOURCES/kvm-block-mirror-Fix-NULL-s-job-in-active-writes.patch
Unescape View File

@ -0,0 +1,75 @@
From b1f5aa5a342a25dc558ee9d435fed0643fe5155f Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Wed, 9 Nov 2022 17:54:50 +0100
Subject: [PATCH 03/11] block/mirror: Fix NULL s->job in active writes
RH-Author: Hanna Czenczek <hreitz@redhat.com>
RH-MergeRequest: 246: block/mirror: Make active mirror progress even under full load
RH-Bugzilla: 2125119
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Stefano Garzarella <sgarzare@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Commit: [3/3] 49d7ebd15667151a6e14228a8260cfdd0aa27a78
There is a small gap in mirror_start_job() before putting the mirror
filter node into the block graph (bdrv_append() call) and the actual job
being created. Before the job is created, MirrorBDSOpaque.job is NULL.
It is possible that requests come in when bdrv_drained_end() is called,
and those requests would see MirrorBDSOpaque.job == NULL. Have our
filter node handle that case gracefully.
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Message-Id: <20221109165452.67927-4-hreitz@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit da93d5c84e56e6b4e84aa8e98b6b984c9b6bb528)
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
---
block/mirror.c | 20 ++++++++++++--------
1 file changed, 12 insertions(+), 8 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 6b02555ad7..50289fca49 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -1438,11 +1438,13 @@ static int coroutine_fn bdrv_mirror_top_do_write(BlockDriverState *bs,
MirrorOp *op = NULL;
MirrorBDSOpaque *s = bs->opaque;
int ret = 0;
- bool copy_to_target;
+ bool copy_to_target = false;
- copy_to_target = s->job->ret >= 0 &&
- !job_is_cancelled(&s->job->common.job) &&
- s->job->copy_mode == MIRROR_COPY_MODE_WRITE_BLOCKING;
+ if (s->job) {
+ copy_to_target = s->job->ret >= 0 &&
+ !job_is_cancelled(&s->job->common.job) &&
+ s->job->copy_mode == MIRROR_COPY_MODE_WRITE_BLOCKING;
+ }
if (copy_to_target) {
op = active_write_prepare(s->job, offset, bytes);
@@ -1487,11 +1489,13 @@ static int coroutine_fn bdrv_mirror_top_pwritev(BlockDriverState *bs,
QEMUIOVector bounce_qiov;
void *bounce_buf;
int ret = 0;
- bool copy_to_target;
+ bool copy_to_target = false;
- copy_to_target = s->job->ret >= 0 &&
- !job_is_cancelled(&s->job->common.job) &&
- s->job->copy_mode == MIRROR_COPY_MODE_WRITE_BLOCKING;
+ if (s->job) {
+ copy_to_target = s->job->ret >= 0 &&
+ !job_is_cancelled(&s->job->common.job) &&
+ s->job->copy_mode == MIRROR_COPY_MODE_WRITE_BLOCKING;
+ }
if (copy_to_target) {
/* The guest might concurrently modify the data to write; but
--
2.37.3

52
SOURCES/kvm-block-nbd-Assert-there-are-no-timers-when-closed.patch
Unescape View File

@ -0,0 +1,52 @@
From d5a85fcf996948d1154e88e9ee3b4e8c64ec2694 Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Fri, 4 Feb 2022 12:10:08 +0100
Subject: [PATCH 2/6] block/nbd: Assert there are no timers when closed
RH-Author: Hanna Reitz <hreitz@redhat.com>
RH-MergeRequest: 117: block/nbd: Handle AioContext changes
RH-Commit: [2/6] 995795ae9844a7d2b28cb1e57fd7fe81482d0205
RH-Bugzilla: 2035185
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Acked-by: Eric Blake <eblake@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Our two timers must not remain armed beyond nbd_clear_bdrvstate(), or
they will access freed data when they fire.
This patch is separate from the patches that actually fix the issue
(HEAD^^ and HEAD^) so that you can run the associated regression iotest
(281) on a configuration that reproducibly exposes the bug.
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
(cherry picked from commit 8a39c381e5e407d2fe5500324323f90a8540fa90)
Conflict:
- block/nbd.c: open_timer was introduced after the 6.2 release (for
nbd's @open-timeout parameter), and has not been backported, so drop
the assertion that it is NULL
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
---
block/nbd.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/block/nbd.c b/block/nbd.c
index b8e5a9b4cc..aab20125d8 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -108,6 +108,9 @@ static void nbd_clear_bdrvstate(BlockDriverState *bs)
yank_unregister_instance(BLOCKDEV_YANK_INSTANCE(bs->node_name));
+ /* Must not leave timers behind that would access freed data */
+ assert(!s->reconnect_delay_timer);
+
object_unref(OBJECT(s->tlscreds));
qapi_free_SocketAddress(s->saddr);
s->saddr = NULL;
--
2.27.0

54
SOURCES/kvm-block-nbd-Delete-reconnect-delay-timer-when-done.patch
Unescape View File

@ -0,0 +1,54 @@
From 8e23c0f208c6bd5bb64c4f6e4863b93fa6f4e9de Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Fri, 4 Feb 2022 12:10:06 +0100
Subject: [PATCH 1/6] block/nbd: Delete reconnect delay timer when done
RH-Author: Hanna Reitz <hreitz@redhat.com>
RH-MergeRequest: 117: block/nbd: Handle AioContext changes
RH-Commit: [1/6] 70814602a8a43a7c14857d76266d82b1aa5174a9
RH-Bugzilla: 2035185
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Acked-by: Eric Blake <eblake@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
We start the reconnect delay timer to cancel the reconnection attempt
after a while. Once nbd_co_do_establish_connection() has returned, this
attempt is over, and we no longer need the timer.
Delete it before returning from nbd_reconnect_attempt(), so that it does
not persist beyond the I/O request that was paused for reconnecting; we
do not want it to fire in a drained section, because all sort of things
can happen in such a section (e.g. the AioContext might be changed, and
we do not want the timer to fire in the wrong context; or the BDS might
even be deleted, and so the timer CB would access already-freed data).
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
(cherry picked from commit 3ce1fc16bad9c3f8b7b10b451a224d6d76e5c551)
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
---
block/nbd.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/block/nbd.c b/block/nbd.c
index 5ef462db1b..b8e5a9b4cc 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -353,6 +353,13 @@ static coroutine_fn void nbd_reconnect_attempt(BDRVNBDState *s)
}
nbd_co_do_establish_connection(s->bs, NULL);
+
+ /*
+ * The reconnect attempt is done (maybe successfully, maybe not), so
+ * we no longer need this timer. Delete it so it will not outlive
+ * this I/O request (so draining removes all timers).
+ */
+ reconnect_delay_timer_del(s);
}
static coroutine_fn int nbd_receive_replies(BDRVNBDState *s, uint64_t handle)
--
2.27.0

107
SOURCES/kvm-block-nbd-Move-s-ioc-on-AioContext-change.patch
Unescape View File

@ -0,0 +1,107 @@
From c7f63e7bbc5119d92775e20d1ebbf8280c78b732 Mon Sep 17 00:00:00 2001
From: Hanna Reitz <hreitz@redhat.com>
Date: Fri, 4 Feb 2022 12:10:11 +0100
Subject: [PATCH 5/6] block/nbd: Move s->ioc on AioContext change
RH-Author: Hanna Reitz <hreitz@redhat.com>
RH-MergeRequest: 117: block/nbd: Handle AioContext changes
RH-Commit: [5/6] 107757b9fbadfb832c75521317108525daa4174e
RH-Bugzilla: 2035185
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Acked-by: Eric Blake <eblake@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
s->ioc must always be attached to the NBD node's AioContext. If that
context changes, s->ioc must be attached to the new context.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2033626
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
(cherry picked from commit e15f3a66c830e3fce99c9d56c493c2f7078a1225)
Conflict:
- block/nbd.c: open_timer was added after the 6.2 release, so we need
not (and cannot) assert it is NULL here.
Signed-off-by: Hanna Reitz <hreitz@redhat.com>
---
block/nbd.c | 41 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 41 insertions(+)
diff --git a/block/nbd.c b/block/nbd.c
index aab20125d8..a3896c7f5f 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -2003,6 +2003,38 @@ static void nbd_cancel_in_flight(BlockDriverState *bs)
nbd_co_establish_connection_cancel(s->conn);
}
+static void nbd_attach_aio_context(BlockDriverState *bs,
+ AioContext *new_context)
+{
+ BDRVNBDState *s = bs->opaque;
+
+ /*
+ * The reconnect_delay_timer is scheduled in I/O paths when the
+ * connection is lost, to cancel the reconnection attempt after a
+ * given time. Once this attempt is done (successfully or not),
+ * nbd_reconnect_attempt() ensures the timer is deleted before the
+ * respective I/O request is resumed.
+ * Since the AioContext can only be changed when a node is drained,
+ * the reconnect_delay_timer cannot be active here.
+ */
+ assert(!s->reconnect_delay_timer);
+
+ if (s->ioc) {
+ qio_channel_attach_aio_context(s->ioc, new_context);
+ }
+}
+
+static void nbd_detach_aio_context(BlockDriverState *bs)
+{
+ BDRVNBDState *s = bs->opaque;
+
+ assert(!s->reconnect_delay_timer);
+
+ if (s->ioc) {
+ qio_channel_detach_aio_context(s->ioc);
+ }
+}
+
static BlockDriver bdrv_nbd = {
.format_name = "nbd",
.protocol_name = "nbd",
@@ -2026,6 +2058,9 @@ static BlockDriver bdrv_nbd = {
.bdrv_dirname = nbd_dirname,
.strong_runtime_opts = nbd_strong_runtime_opts,
.bdrv_cancel_in_flight = nbd_cancel_in_flight,
+
+ .bdrv_attach_aio_context = nbd_attach_aio_context,
+ .bdrv_detach_aio_context = nbd_detach_aio_context,
};
static BlockDriver bdrv_nbd_tcp = {
@@ -2051,6 +2086,9 @@ static BlockDriver bdrv_nbd_tcp = {
.bdrv_dirname = nbd_dirname,
.strong_runtime_opts = nbd_strong_runtime_opts,
.bdrv_cancel_in_flight = nbd_cancel_in_flight,
+
+ .bdrv_attach_aio_context = nbd_attach_aio_context,
+ .bdrv_detach_aio_context = nbd_detach_aio_context,
};
static BlockDriver bdrv_nbd_unix = {
@@ -2076,6 +2114,9 @@ static BlockDriver bdrv_nbd_unix = {
.bdrv_dirname = nbd_dirname,
.strong_runtime_opts = nbd_strong_runtime_opts,
.bdrv_cancel_in_flight = nbd_cancel_in_flight,
+
+ .bdrv_attach_aio_context = nbd_attach_aio_context,
+ .bdrv_detach_aio_context = nbd_detach_aio_context,
};
static void bdrv_nbd_init(void)
--
2.27.0

59
SOURCES/kvm-block-rbd-fix-handling-of-holes-in-.bdrv_co_block_st.patch
Unescape View File

@ -0,0 +1,59 @@
From f4b7133d7aeb1d0b9115d01b5cff4df7f6b24e78 Mon Sep 17 00:00:00 2001
From: Peter Lieven <pl@kamp.de>
Date: Thu, 13 Jan 2022 15:44:25 +0100
Subject: [PATCH 5/6] block/rbd: fix handling of holes in .bdrv_co_block_status
RH-Author: Stefano Garzarella <sgarzare@redhat.com>
RH-MergeRequest: 110: block/rbd: fix handling of holes in .bdrv_co_block_status
RH-Commit: [1/2] 352656a5c77cc7855b476c3559a10c6aa64a4f58
RH-Bugzilla: 2037135
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Acked-by: Hanna Reitz <hreitz@redhat.com>
the assumption that we can't hit a hole if we do not diff against a snapshot was wrong.
We can see a hole in an image if we diff against base if there exists an older snapshot
of the image and we have discarded blocks in the image where the snapshot has data.
Fix this by simply handling a hole like an unallocated area. There are no callbacks
for unallocated areas so just bail out if we hit a hole.
Fixes: 0347a8fd4c3faaedf119be04c197804be40a384b
Suggested-by: Ilya Dryomov <idryomov@gmail.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Peter Lieven <pl@kamp.de>
Message-Id: <20220113144426.4036493-2-pl@kamp.de>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit 9e302f64bb407a9bb097b626da97228c2654cfee)
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
block/rbd.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/block/rbd.c b/block/rbd.c
index def96292e0..20bb896c4a 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1279,11 +1279,11 @@ static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
RBDDiffIterateReq *req = opaque;
assert(req->offs + req->bytes <= offs);
- /*
- * we do not diff against a snapshot so we should never receive a callback
- * for a hole.
- */
- assert(exists);
+
+ /* treat a hole like an unallocated area and bail out */
+ if (!exists) {
+ return 0;
+ }
if (!req->exists && offs > req->offs) {
/*
--
2.27.0

103
SOURCES/kvm-block-rbd-workaround-for-ceph-issue-53784.patch
Unescape View File

@ -0,0 +1,103 @@
From 8c50eedf03d8e62acd387b9aa9369dadcea9324c Mon Sep 17 00:00:00 2001
From: Peter Lieven <pl@kamp.de>
Date: Thu, 13 Jan 2022 15:44:26 +0100
Subject: [PATCH 6/6] block/rbd: workaround for ceph issue #53784
RH-Author: Stefano Garzarella <sgarzare@redhat.com>
RH-MergeRequest: 110: block/rbd: fix handling of holes in .bdrv_co_block_status
RH-Commit: [2/2] 1384557462e89bb539d0d25a1a471ad738fb9e89
RH-Bugzilla: 2037135
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Acked-by: Hanna Reitz <hreitz@redhat.com>
librbd had a bug until early 2022 that affected all versions of ceph that
supported fast-diff. This bug results in reporting of incorrect offsets
if the offset parameter to rbd_diff_iterate2 is not object aligned.
This patch works around this bug for pre Quincy versions of librbd.
Fixes: 0347a8fd4c3faaedf119be04c197804be40a384b
Cc: qemu-stable@nongnu.org
Signed-off-by: Peter Lieven <pl@kamp.de>
Message-Id: <20220113144426.4036493-3-pl@kamp.de>
Reviewed-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Tested-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit fc176116cdea816ceb8dd969080b2b95f58edbc0)
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
block/rbd.c | 42 ++++++++++++++++++++++++++++++++++++++++--
1 file changed, 40 insertions(+), 2 deletions(-)
diff --git a/block/rbd.c b/block/rbd.c
index 20bb896c4a..8f183eba2a 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1320,6 +1320,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
int status, r;
RBDDiffIterateReq req = { .offs = offset };
uint64_t features, flags;
+ uint64_t head = 0;
assert(offset + bytes <= s->image_size);
@@ -1347,7 +1348,43 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
return status;
}
- r = rbd_diff_iterate2(s->image, NULL, offset, bytes, true, true,
+#if LIBRBD_VERSION_CODE < LIBRBD_VERSION(1, 17, 0)
+ /*
+ * librbd had a bug until early 2022 that affected all versions of ceph that
+ * supported fast-diff. This bug results in reporting of incorrect offsets
+ * if the offset parameter to rbd_diff_iterate2 is not object aligned.
+ * Work around this bug by rounding down the offset to object boundaries.
+ * This is OK because we call rbd_diff_iterate2 with whole_object = true.
+ * However, this workaround only works for non cloned images with default
+ * striping.
+ *
+ * See: https://tracker.ceph.com/issues/53784
+ */
+
+ /* check if RBD image has non-default striping enabled */
+ if (features & RBD_FEATURE_STRIPINGV2) {
+ return status;
+ }
+
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wdeprecated-declarations"
+ /*
+ * check if RBD image is a clone (= has a parent).
+ *
+ * rbd_get_parent_info is deprecated from Nautilus onwards, but the
+ * replacement rbd_get_parent is not present in Luminous and Mimic.
+ */
+ if (rbd_get_parent_info(s->image, NULL, 0, NULL, 0, NULL, 0) != -ENOENT) {
+ return status;
+ }
+#pragma GCC diagnostic pop
+
+ head = req.offs & (s->object_size - 1);
+ req.offs -= head;
+ bytes += head;
+#endif
+
+ r = rbd_diff_iterate2(s->image, NULL, req.offs, bytes, true, true,
qemu_rbd_diff_iterate_cb, &req);
if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
return status;
@@ -1366,7 +1403,8 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
}
- *pnum = req.bytes;
+ assert(req.bytes > head);
+ *pnum = req.bytes - head;
return status;
}
--
2.27.0

58
SOURCES/kvm-display-qxl-render-fix-race-condition-in-qxl_cursor-.patch
Unescape View File

@ -0,0 +1,58 @@
From abd84f26e0fe0bc9952d91fbd35fb3a7253cfecf Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@rehat.com>
Date: Wed, 13 Apr 2022 20:54:45 -0400
Subject: [PATCH 1/2] display/qxl-render: fix race condition in qxl_cursor
(CVE-2021-4207)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 152: display/qxl-render: fix race condition in qxl_cursor (CVE-2021-4207)
RH-Commit: [1/1] f05b9a956f2e0ca522b5be127beff813d04b5588 (jmaloy/qemu-kvm)
RH-Bugzilla: 2040738
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Mauro Matteo Cascella <None>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2040738
Upstream: Merged
CVE: CVE-2021-4207
commit 9569f5cb5b4bffa9d3ebc8ba7da1e03830a9a895
Author: Mauro Matteo Cascella <mcascell@redhat.com>
Date: Thu Apr 7 10:11:06 2022 +0200
display/qxl-render: fix race condition in qxl_cursor (CVE-2021-4207)
Avoid fetching 'width' and 'height' a second time to prevent possible
race condition. Refer to security advisory
https://starlabs.sg/advisories/22-4207/ for more information.
Fixes: CVE-2021-4207
Signed-off-by: Mauro Matteo Cascella <mcascell@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220407081106.343235-1-mcascell@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 9569f5cb5b4bffa9d3ebc8ba7da1e03830a9a895)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/display/qxl-render.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/display/qxl-render.c b/hw/display/qxl-render.c
index d28849b121..237ed293ba 100644
--- a/hw/display/qxl-render.c
+++ b/hw/display/qxl-render.c
@@ -266,7 +266,7 @@ static QEMUCursor *qxl_cursor(PCIQXLDevice *qxl, QXLCursor *cursor,
}
break;
case SPICE_CURSOR_TYPE_ALPHA:
- size = sizeof(uint32_t) * cursor->header.width * cursor->header.height;
+ size = sizeof(uint32_t) * c->width * c->height;
qxl_unpack_chunks(c->data, size, qxl, &cursor->chunk, group_id);
if (qxl->debug > 2) {
cursor_print_ascii_art(c, "qxl/alpha");
--
2.27.0

127
SOURCES/kvm-dma-helpers-prevent-dma_blk_cb-vs-dma_aio_cancel-rac.patch
Unescape View File

@ -0,0 +1,127 @@
From 103608465b8bd2edf7f9aaef5c3c93309ccf9ec2 Mon Sep 17 00:00:00 2001
From: Stefan Hajnoczi <stefanha@redhat.com>
Date: Tue, 21 Feb 2023 16:22:17 -0500
Subject: [PATCH 12/13] dma-helpers: prevent dma_blk_cb() vs dma_aio_cancel()
race
RH-Author: Stefan Hajnoczi <stefanha@redhat.com>
RH-MergeRequest: 264: scsi: protect req->aiocb with AioContext lock
RH-Bugzilla: 2090990
RH-Acked-by: Stefano Garzarella <sgarzare@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com>
RH-Acked-by: Kevin Wolf <kwolf@redhat.com>
RH-Commit: [2/3] 14f5835093ba8c5111f3ada2fe87730371aca733
dma_blk_cb() only takes the AioContext lock around ->io_func(). That
means the rest of dma_blk_cb() is not protected. In particular, the
DMAAIOCB field accesses happen outside the lock.
There is a race when the main loop thread holds the AioContext lock and
invokes scsi_device_purge_requests() -> bdrv_aio_cancel() ->
dma_aio_cancel() while an IOThread executes dma_blk_cb(). The dbs->acb
field determines how cancellation proceeds. If dma_aio_cancel() sees
dbs->acb == NULL while dma_blk_cb() is still running, the request can be
completed twice (-ECANCELED and the actual return value).
The following assertion can occur with virtio-scsi when an IOThread is
used:
../hw/scsi/scsi-disk.c:368: scsi_dma_complete: Assertion `r->req.aiocb != NULL' failed.
Fix the race by holding the AioContext across dma_blk_cb(). Now
dma_aio_cancel() under the AioContext lock will not see
inconsistent/intermediate states.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20230221212218.1378734-3-stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
(cherry picked from commit abfcd2760b3e70727bbc0792221b8b98a733dc32)
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
hw/scsi/scsi-disk.c | 4 +---
softmmu/dma-helpers.c | 12 +++++++-----
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c
index 179ce22c4a..c8109a673e 100644
--- a/hw/scsi/scsi-disk.c
+++ b/hw/scsi/scsi-disk.c
@@ -351,13 +351,12 @@ done:
scsi_req_unref(&r->req);
}
+/* Called with AioContext lock held */
static void scsi_dma_complete(void *opaque, int ret)
{
SCSIDiskReq *r = (SCSIDiskReq *)opaque;
SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
- aio_context_acquire(blk_get_aio_context(s->qdev.conf.blk));
-
assert(r->req.aiocb != NULL);
r->req.aiocb = NULL;
@@ -367,7 +366,6 @@ static void scsi_dma_complete(void *opaque, int ret)
block_acct_done(blk_get_stats(s->qdev.conf.blk), &r->acct);
}
scsi_dma_complete_noio(r, ret);
- aio_context_release(blk_get_aio_context(s->qdev.conf.blk));
}
static void scsi_read_complete_noio(SCSIDiskReq *r, int ret)
diff --git a/softmmu/dma-helpers.c b/softmmu/dma-helpers.c
index 7d766a5e89..42af18719a 100644
--- a/softmmu/dma-helpers.c
+++ b/softmmu/dma-helpers.c
@@ -127,17 +127,19 @@ static void dma_complete(DMAAIOCB *dbs, int ret)
static void dma_blk_cb(void *opaque, int ret)
{
DMAAIOCB *dbs = (DMAAIOCB *)opaque;
+ AioContext *ctx = dbs->ctx;
dma_addr_t cur_addr, cur_len;
void *mem;
trace_dma_blk_cb(dbs, ret);
+ aio_context_acquire(ctx);
dbs->acb = NULL;
dbs->offset += dbs->iov.size;
if (dbs->sg_cur_index == dbs->sg->nsg || ret < 0) {
dma_complete(dbs, ret);
- return;
+ goto out;
}
dma_blk_unmap(dbs);
@@ -177,9 +179,9 @@ static void dma_blk_cb(void *opaque, int ret)
if (dbs->iov.size == 0) {
trace_dma_map_wait(dbs);
- dbs->bh = aio_bh_new(dbs->ctx, reschedule_dma, dbs);
+ dbs->bh = aio_bh_new(ctx, reschedule_dma, dbs);
cpu_register_map_client(dbs->bh);
- return;
+ goto out;
}
if (!QEMU_IS_ALIGNED(dbs->iov.size, dbs->align)) {
@@ -187,11 +189,11 @@ static void dma_blk_cb(void *opaque, int ret)
QEMU_ALIGN_DOWN(dbs->iov.size, dbs->align));
}
- aio_context_acquire(dbs->ctx);
dbs->acb = dbs->io_func(dbs->offset, &dbs->iov,
dma_blk_cb, dbs, dbs->io_func_opaque);
- aio_context_release(dbs->ctx);
assert(dbs->acb);
+out:
+ aio_context_release(ctx);
}
static void dma_aio_cancel(BlockAIOCB *acb)
--
2.37.3

77
SOURCES/kvm-doc-Add-the-SGX-numa-description.patch
Unescape View File

@ -0,0 +1,77 @@
From e8377e3f4d540e2594a50985523e87d1f3cabbc7 Mon Sep 17 00:00:00 2001
From: Yang Zhong <yang.zhong@intel.com>
Date: Mon, 1 Nov 2021 12:20:08 -0400
Subject: [PATCH 3/7] doc: Add the SGX numa description
RH-Author: Paul Lai <None>
RH-MergeRequest: 111: numa: Enable numa for SGX EPC sections
RH-Commit: [3/5] 41c74688c9662b966c243566a837135ff52341c4
RH-Bugzilla: 1518984
RH-Acked-by: Paolo Bonzini <None>
RH-Acked-by: Bandan Das <None>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
Add the SGX numa reference command and how to check if
SGX numa is support or not with multiple EPC sections.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20211101162009.62161-5-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit d1889b36098c79e2e6ac90faf3d0dc5ec0057677)
Signed-off-by: Paul Lai <plai@redhat.com>
---
docs/system/i386/sgx.rst | 31 +++++++++++++++++++++++++++----
1 file changed, 27 insertions(+), 4 deletions(-)
diff --git a/docs/system/i386/sgx.rst b/docs/system/i386/sgx.rst
index f8fade5ac2..0f0a73f758 100644
--- a/docs/system/i386/sgx.rst
+++ b/docs/system/i386/sgx.rst
@@ -141,8 +141,7 @@ To launch a SGX guest:
|qemu_system_x86| \\
-cpu host,+sgx-provisionkey \\
-object memory-backend-epc,id=mem1,size=64M,prealloc=on \\
- -object memory-backend-epc,id=mem2,size=28M \\
- -M sgx-epc.0.memdev=mem1,sgx-epc.1.memdev=mem2
+ -M sgx-epc.0.memdev=mem1,sgx-epc.0.node=0
Utilizing SGX in the guest requires a kernel/OS with SGX support.
The support can be determined in guest by::
@@ -152,8 +151,32 @@ The support can be determined in guest by::
and SGX epc info by::
$ dmesg | grep sgx
- [ 1.242142] sgx: EPC section 0x180000000-0x181bfffff
- [ 1.242319] sgx: EPC section 0x181c00000-0x1837fffff
+ [ 0.182807] sgx: EPC section 0x140000000-0x143ffffff
+ [ 0.183695] sgx: [Firmware Bug]: Unable to map EPC section to online node. Fallback to the NUMA node 0.
+
+To launch a SGX numa guest:
+
+.. parsed-literal::
+
+ |qemu_system_x86| \\
+ -cpu host,+sgx-provisionkey \\
+ -object memory-backend-ram,size=2G,host-nodes=0,policy=bind,id=node0 \\
+ -object memory-backend-epc,id=mem0,size=64M,prealloc=on,host-nodes=0,policy=bind \\
+ -numa node,nodeid=0,cpus=0-1,memdev=node0 \\
+ -object memory-backend-ram,size=2G,host-nodes=1,policy=bind,id=node1 \\
+ -object memory-backend-epc,id=mem1,size=28M,prealloc=on,host-nodes=1,policy=bind \\
+ -numa node,nodeid=1,cpus=2-3,memdev=node1 \\
+ -M sgx-epc.0.memdev=mem0,sgx-epc.0.node=0,sgx-epc.1.memdev=mem1,sgx-epc.1.node=1
+
+and SGX epc numa info by::
+
+ $ dmesg | grep sgx
+ [ 0.369937] sgx: EPC section 0x180000000-0x183ffffff
+ [ 0.370259] sgx: EPC section 0x184000000-0x185bfffff
+
+ $ dmesg | grep SRAT
+ [ 0.009981] ACPI: SRAT: Node 0 PXM 0 [mem 0x180000000-0x183ffffff]
+ [ 0.009982] ACPI: SRAT: Node 1 PXM 1 [mem 0x184000000-0x185bfffff]
References
----------
--
2.27.0

70
SOURCES/kvm-docs-system-s390x-Document-the-loadparm-machine-prop.patch
Unescape View File

@ -0,0 +1,70 @@
From 407e23d7f0c9020404247afe7d4df98505222bbb Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Mon, 14 Nov 2022 14:25:02 +0100
Subject: [PATCH 1/3] docs/system/s390x: Document the "loadparm" machine
property
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Thomas Huth <thuth@redhat.com>
RH-MergeRequest: 233: s390x: Document the "loadparm" machine property
RH-Bugzilla: 2128225
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Cédric Le Goater <clg@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [1/2] e9589ea32d2a8f82971476b644e1063fa14cf822
The "loadparm" machine property is useful for selecting alternative
kernels on the disk of the guest, but so far we do not tell the users
yet how to use it. Add some documentation to fill this gap.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2128235
Message-Id: <20221114132502.110213-1-thuth@redhat.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit be5df2edb5d69ff3107c5616aa035a9ba8d0422e)
---
docs/system/s390x/bootdevices.rst | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/docs/system/s390x/bootdevices.rst b/docs/system/s390x/bootdevices.rst
index 9e591cb9dc..d4bf3b9f0b 100644
--- a/docs/system/s390x/bootdevices.rst
+++ b/docs/system/s390x/bootdevices.rst
@@ -53,6 +53,32 @@ recommended to specify a CD-ROM device via ``-device scsi-cd`` (as mentioned
above) instead.
+Selecting kernels with the ``loadparm`` property
+------------------------------------------------
+
+The ``s390-ccw-virtio`` machine supports the so-called ``loadparm`` parameter
+which can be used to select the kernel on the disk of the guest that the
+s390-ccw bios should boot. When starting QEMU, it can be specified like this::
+
+ qemu-system-s390x -machine s390-ccw-virtio,loadparm=<string>
+
+The first way to use this parameter is to use the word ``PROMPT`` as the
+``<string>`` here. In that case the s390-ccw bios will show a list of
+installed kernels on the disk of the guest and ask the user to enter a number
+to chose which kernel should be booted -- similar to what can be achieved by
+specifying the ``-boot menu=on`` option when starting QEMU. Note that the menu
+list will only show the names of the installed kernels when using a DASD-like
+disk image with 4k byte sectors. On normal SCSI-style disks with 512-byte
+sectors, there is not enough space for the zipl loader on the disk to store
+the kernel names, so you only get a list without names here.
+
+The second way to use this parameter is to use a number in the range from 0
+to 31. The numbers that can be used here correspond to the numbers that are
+shown when using the ``PROMPT`` option, and the s390-ccw bios will then try
+to automatically boot the kernel that is associated with the given number.
+Note that ``0`` can be used to boot the default entry.
+
+
Booting from a network device
-----------------------------
--
2.37.3

356
SOURCES/kvm-dump-Add-architecture-section-and-section-string-tab.patch
Unescape View File

@ -0,0 +1,356 @@
From f2f3efff83dddd38a97699cd2701f46f61a732e3 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 11:32:10 +0000
Subject: [PATCH 36/42] dump: Add architecture section and section string table
support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [36/41] 83b98ff185e93e62703f686b65546d60c783d783
Add hooks which architectures can use to add arbitrary data to custom
sections.
Also add a section name string table in order to identify section
contents
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017113210.41674-1-frankja@linux.ibm.com>
(cherry picked from commit 9b72224f44612ddd5b434a1bccf79346946d11da)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 186 +++++++++++++++++++++++++++++++------
include/sysemu/dump-arch.h | 3 +
include/sysemu/dump.h | 3 +
3 files changed, 166 insertions(+), 26 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 7a42401790..4aa8fb64d2 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -104,6 +104,7 @@ static int dump_cleanup(DumpState *s)
memory_mapping_list_free(&s->list);
close(s->fd);
g_free(s->guest_note);
+ g_array_unref(s->string_table_buf);
s->guest_note = NULL;
if (s->resume) {
if (s->detached) {
@@ -153,11 +154,10 @@ static void prepare_elf64_header(DumpState *s, Elf64_Ehdr *elf_header)
elf_header->e_phoff = cpu_to_dump64(s, s->phdr_offset);
elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
elf_header->e_phnum = cpu_to_dump16(s, phnum);
- if (s->shdr_num) {
- elf_header->e_shoff = cpu_to_dump64(s, s->shdr_offset);
- elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
- elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
- }
+ elf_header->e_shoff = cpu_to_dump64(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
+ elf_header->e_shstrndx = cpu_to_dump16(s, s->shdr_num - 1);
}
static void prepare_elf32_header(DumpState *s, Elf32_Ehdr *elf_header)
@@ -181,11 +181,10 @@ static void prepare_elf32_header(DumpState *s, Elf32_Ehdr *elf_header)
elf_header->e_phoff = cpu_to_dump32(s, s->phdr_offset);
elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
elf_header->e_phnum = cpu_to_dump16(s, phnum);
- if (s->shdr_num) {
- elf_header->e_shoff = cpu_to_dump32(s, s->shdr_offset);
- elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
- elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
- }
+ elf_header->e_shoff = cpu_to_dump32(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
+ elf_header->e_shstrndx = cpu_to_dump16(s, s->shdr_num - 1);
}
static void write_elf_header(DumpState *s, Error **errp)
@@ -196,6 +195,8 @@ static void write_elf_header(DumpState *s, Error **errp)
void *header_ptr;
int ret;
+ /* The NULL header and the shstrtab are always defined */
+ assert(s->shdr_num >= 2);
if (dump_is_64bit(s)) {
prepare_elf64_header(s, &elf64_header);
header_size = sizeof(elf64_header);
@@ -394,17 +395,49 @@ static void prepare_elf_section_hdr_zero(DumpState *s)
}
}
-static void prepare_elf_section_hdrs(DumpState *s)
+static void prepare_elf_section_hdr_string(DumpState *s, void *buff)
+{
+ uint64_t index = s->string_table_buf->len;
+ const char strtab[] = ".shstrtab";
+ Elf32_Shdr shdr32 = {};
+ Elf64_Shdr shdr64 = {};
+ int shdr_size;
+ void *shdr;
+
+ g_array_append_vals(s->string_table_buf, strtab, sizeof(strtab));
+ if (dump_is_64bit(s)) {
+ shdr_size = sizeof(Elf64_Shdr);
+ shdr64.sh_type = SHT_STRTAB;
+ shdr64.sh_offset = s->section_offset + s->elf_section_data_size;
+ shdr64.sh_name = index;
+ shdr64.sh_size = s->string_table_buf->len;
+ shdr = &shdr64;
+ } else {
+ shdr_size = sizeof(Elf32_Shdr);
+ shdr32.sh_type = SHT_STRTAB;
+ shdr32.sh_offset = s->section_offset + s->elf_section_data_size;
+ shdr32.sh_name = index;
+ shdr32.sh_size = s->string_table_buf->len;
+ shdr = &shdr32;
+ }
+ memcpy(buff, shdr, shdr_size);
+}
+
+static bool prepare_elf_section_hdrs(DumpState *s, Error **errp)
{
size_t len, sizeof_shdr;
+ void *buff_hdr;
/*
* Section ordering:
* - HDR zero
+ * - Arch section hdrs
+ * - String table hdr
*/
sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
len = sizeof_shdr * s->shdr_num;
s->elf_section_hdrs = g_malloc0(len);
+ buff_hdr = s->elf_section_hdrs;
/*
* The first section header is ALWAYS a special initial section
@@ -420,6 +453,26 @@ static void prepare_elf_section_hdrs(DumpState *s)
if (s->phdr_num >= PN_XNUM) {
prepare_elf_section_hdr_zero(s);
}
+ buff_hdr += sizeof_shdr;
+
+ /* Add architecture defined section headers */
+ if (s->dump_info.arch_sections_write_hdr_fn
+ && s->shdr_num > 2) {
+ buff_hdr += s->dump_info.arch_sections_write_hdr_fn(s, buff_hdr);
+
+ if (s->shdr_num >= SHN_LORESERVE) {
+ error_setg_errno(errp, EINVAL,
+ "dump: too many architecture defined sections");
+ return false;
+ }
+ }
+
+ /*
+ * String table is the last section since strings are added via
+ * arch_sections_write_hdr().
+ */
+ prepare_elf_section_hdr_string(s, buff_hdr);
+ return true;
}
static void write_elf_section_headers(DumpState *s, Error **errp)
@@ -427,7 +480,9 @@ static void write_elf_section_headers(DumpState *s, Error **errp)
size_t sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
int ret;
- prepare_elf_section_hdrs(s);
+ if (!prepare_elf_section_hdrs(s, errp)) {
+ return;
+ }
ret = fd_write_vmcore(s->elf_section_hdrs, s->shdr_num * sizeof_shdr, s);
if (ret < 0) {
@@ -437,6 +492,29 @@ static void write_elf_section_headers(DumpState *s, Error **errp)
g_free(s->elf_section_hdrs);
}
+static void write_elf_sections(DumpState *s, Error **errp)
+{
+ int ret;
+
+ if (s->elf_section_data_size) {
+ /* Write architecture section data */
+ ret = fd_write_vmcore(s->elf_section_data,
+ s->elf_section_data_size, s);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "dump: failed to write architecture section data");
+ return;
+ }
+ }
+
+ /* Write string table */
+ ret = fd_write_vmcore(s->string_table_buf->data,
+ s->string_table_buf->len, s);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "dump: failed to write string table data");
+ }
+}
+
static void write_data(DumpState *s, void *buf, int length, Error **errp)
{
int ret;
@@ -693,6 +771,31 @@ static void dump_iterate(DumpState *s, Error **errp)
}
}
+static void dump_end(DumpState *s, Error **errp)
+{
+ int rc;
+ ERRP_GUARD();
+
+ if (s->elf_section_data_size) {
+ s->elf_section_data = g_malloc0(s->elf_section_data_size);
+ }
+
+ /* Adds the architecture defined section data to s->elf_section_data */
+ if (s->dump_info.arch_sections_write_fn &&
+ s->elf_section_data_size) {
+ rc = s->dump_info.arch_sections_write_fn(s, s->elf_section_data);
+ if (rc) {
+ error_setg_errno(errp, rc,
+ "dump: failed to get arch section data");
+ g_free(s->elf_section_data);
+ return;
+ }
+ }
+
+ /* write sections to vmcore */
+ write_elf_sections(s, errp);
+}
+
static void create_vmcore(DumpState *s, Error **errp)
{
ERRP_GUARD();
@@ -702,7 +805,14 @@ static void create_vmcore(DumpState *s, Error **errp)
return;
}
+ /* Iterate over memory and dump it to file */
dump_iterate(s, errp);
+ if (*errp) {
+ return;
+ }
+
+ /* Write the section data */
+ dump_end(s, errp);
}
static int write_start_flat_header(int fd)
@@ -1720,6 +1830,14 @@ static void dump_init(DumpState *s, int fd, bool has_format,
s->filter_area_begin = begin;
s->filter_area_length = length;
+ /* First index is 0, it's the special null name */
+ s->string_table_buf = g_array_new(FALSE, TRUE, 1);
+ /*
+ * Allocate the null name, due to the clearing option set to true
+ * it will be 0.
+ */
+ g_array_set_size(s->string_table_buf, 1);
+
memory_mapping_list_init(&s->list);
guest_phys_blocks_init(&s->guest_phys_blocks);
@@ -1856,26 +1974,42 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
/*
- * calculate phdr_num
+ * The first section header is always a special one in which most
+ * fields are 0. The section header string table is also always
+ * set.
+ */
+ s->shdr_num = 2;
+
+ /*
+ * Adds the number of architecture sections to shdr_num and sets
+ * elf_section_data_size so we know the offsets and sizes of all
+ * parts.
+ */
+ if (s->dump_info.arch_sections_add_fn) {
+ s->dump_info.arch_sections_add_fn(s);
+ }
+
+ /*
+ * calculate shdr_num so we know the offsets and sizes of all
+ * parts.
+ * Calculate phdr_num
*
- * the type of ehdr->e_phnum is uint16_t, so we should avoid overflow
+ * The absolute maximum amount of phdrs is UINT32_MAX - 1 as
+ * sh_info is 32 bit. There's special handling once we go over
+ * UINT16_MAX - 1 but that is handled in the ehdr and section
+ * code.
*/
- s->phdr_num = 1; /* PT_NOTE */
- if (s->list.num < UINT16_MAX - 2) {
- s->shdr_num = 0;
+ s->phdr_num = 1; /* Reserve PT_NOTE */
+ if (s->list.num <= UINT32_MAX - 1) {
s->phdr_num += s->list.num;
} else {
- /* sh_info of section 0 holds the real number of phdrs */
- s->shdr_num = 1;
-
- /* the type of shdr->sh_info is uint32_t, so we should avoid overflow */
- if (s->list.num <= UINT32_MAX - 1) {
- s->phdr_num += s->list.num;
- } else {
- s->phdr_num = UINT32_MAX;
- }
+ s->phdr_num = UINT32_MAX;
}
+ /*
+ * Now that the number of section and program headers is known we
+ * can calculate the offsets of the headers and data.
+ */
if (dump_is_64bit(s)) {
s->shdr_offset = sizeof(Elf64_Ehdr);
s->phdr_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
diff --git a/include/sysemu/dump-arch.h b/include/sysemu/dump-arch.h
index e25b02e990..59bbc9be38 100644
--- a/include/sysemu/dump-arch.h
+++ b/include/sysemu/dump-arch.h
@@ -21,6 +21,9 @@ typedef struct ArchDumpInfo {
uint32_t page_size; /* The target's page size. If it's variable and
* unknown, then this should be the maximum. */
uint64_t phys_base; /* The target's physmem base. */
+ void (*arch_sections_add_fn)(DumpState *s);
+ uint64_t (*arch_sections_write_hdr_fn)(DumpState *s, uint8_t *buff);
+ int (*arch_sections_write_fn)(DumpState *s, uint8_t *buff);
} ArchDumpInfo;
struct GuestPhysBlockList; /* memory_mapping.h */
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 9ed811b313..38ccac7190 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -180,6 +180,9 @@ typedef struct DumpState {
hwaddr note_offset;
void *elf_section_hdrs; /* Pointer to section header buffer */
+ void *elf_section_data; /* Pointer to section data buffer */
+ uint64_t elf_section_data_size; /* Size of section data */
+ GArray *string_table_buf; /* String table data buffer */
uint8_t *note_buf; /* buffer for notes */
size_t note_buf_offset; /* the writing place in note_buf */
--
2.37.3

138
SOURCES/kvm-dump-Add-more-offset-variables.patch
Unescape View File

@ -0,0 +1,138 @@
From bee31226b87d0b05faae84e88cce3af1b8dabbfd Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:59 +0000
Subject: [PATCH 17/42] dump: Add more offset variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [17/41] fbe629e1476e8a0e039f989af6e1f4707075ba01
Offset calculations are easy enough to get wrong. Let's add a few
variables to make moving around elf headers and data sections easier.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20220330123603.107120-6-frankja@linux.ibm.com>
(cherry picked from commit e71d353360bb09a8e784e35d78370c691f6ea185)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 35 +++++++++++++++--------------------
include/sysemu/dump.h | 4 ++++
2 files changed, 19 insertions(+), 20 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 5cc2322325..85a402b38c 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -142,13 +142,11 @@ static void write_elf64_header(DumpState *s, Error **errp)
elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump64(s, sizeof(Elf64_Ehdr));
+ elf_header.e_phoff = cpu_to_dump64(s, s->phdr_offset);
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->phdr_num;
-
- elf_header.e_shoff = cpu_to_dump64(s, shoff);
+ elf_header.e_shoff = cpu_to_dump64(s, s->shdr_offset);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
@@ -179,13 +177,11 @@ static void write_elf32_header(DumpState *s, Error **errp)
elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump32(s, sizeof(Elf32_Ehdr));
+ elf_header.e_phoff = cpu_to_dump32(s, s->phdr_offset);
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->phdr_num;
-
- elf_header.e_shoff = cpu_to_dump32(s, shoff);
+ elf_header.e_shoff = cpu_to_dump32(s, s->shdr_offset);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
@@ -248,12 +244,11 @@ static void write_elf32_load(DumpState *s, MemoryMapping *memory_mapping,
static void write_elf64_note(DumpState *s, Error **errp)
{
Elf64_Phdr phdr;
- hwaddr begin = s->memory_offset - s->note_size;
int ret;
memset(&phdr, 0, sizeof(Elf64_Phdr));
phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump64(s, begin);
+ phdr.p_offset = cpu_to_dump64(s, s->note_offset);
phdr.p_paddr = 0;
phdr.p_filesz = cpu_to_dump64(s, s->note_size);
phdr.p_memsz = cpu_to_dump64(s, s->note_size);
@@ -313,13 +308,12 @@ static void write_elf64_notes(WriteCoreDumpFunction f, DumpState *s,
static void write_elf32_note(DumpState *s, Error **errp)
{
- hwaddr begin = s->memory_offset - s->note_size;
Elf32_Phdr phdr;
int ret;
memset(&phdr, 0, sizeof(Elf32_Phdr));
phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump32(s, begin);
+ phdr.p_offset = cpu_to_dump32(s, s->note_offset);
phdr.p_paddr = 0;
phdr.p_filesz = cpu_to_dump32(s, s->note_size);
phdr.p_memsz = cpu_to_dump32(s, s->note_size);
@@ -1826,15 +1820,16 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (s->dump_info.d_class == ELFCLASS64) {
- s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->phdr_num +
- sizeof(Elf64_Shdr) * s->shdr_num +
- s->note_size;
+ s->phdr_offset = sizeof(Elf64_Ehdr);
+ s->shdr_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
+ s->note_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
+ s->memory_offset = s->note_offset + s->note_size;
} else {
- s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->phdr_num +
- sizeof(Elf32_Shdr) * s->shdr_num +
- s->note_size;
+
+ s->phdr_offset = sizeof(Elf32_Ehdr);
+ s->shdr_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
+ s->note_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
+ s->memory_offset = s->note_offset + s->note_size;
}
return;
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 19458bffbd..ffc2ea1072 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -159,6 +159,10 @@ typedef struct DumpState {
bool resume;
bool detached;
ssize_t note_size;
+ hwaddr shdr_offset;
+ hwaddr phdr_offset;
+ hwaddr section_offset;
+ hwaddr note_offset;
hwaddr memory_offset;
int fd;
--
2.37.3

94
SOURCES/kvm-dump-Cleanup-dump_begin-write-functions.patch
Unescape View File

@ -0,0 +1,94 @@
From cbb653d73e32513ccd46b293a52384eed6a5f84f Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:02 +0000
Subject: [PATCH 20/42] dump: Cleanup dump_begin write functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [20/41] 18ea1457a3e54fd368e556d96c3be50c6ad0a6bd
There's no need to have a gigantic if in there let's move the elf
32/64 bit logic into the section, segment or note code.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-9-frankja@linux.ibm.com>
(cherry picked from commit 5ff2e5a3e1e67930e523486e39549a33fcf97227)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 42 +++++++++++-------------------------------
1 file changed, 11 insertions(+), 31 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 823ca32883..88abde355a 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -565,46 +565,26 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- if (dump_is_64bit(s)) {
- /* write all PT_LOAD to vmcore */
- write_elf_loads(s, errp);
+ /* write all PT_LOAD to vmcore */
+ write_elf_loads(s, errp);
+ if (*errp) {
+ return;
+ }
+
+ /* write section to vmcore */
+ if (s->shdr_num) {
+ write_elf_section(s, 1, errp);
if (*errp) {
return;
}
+ }
- /* write section to vmcore */
- if (s->shdr_num) {
- write_elf_section(s, 1, errp);
- if (*errp) {
- return;
- }
- }
-
+ if (dump_is_64bit(s)) {
/* write notes to vmcore */
write_elf64_notes(fd_write_vmcore, s, errp);
- if (*errp) {
- return;
- }
} else {
- /* write all PT_LOAD to vmcore */
- write_elf_loads(s, errp);
- if (*errp) {
- return;
- }
-
- /* write section to vmcore */
- if (s->shdr_num) {
- write_elf_section(s, 0, errp);
- if (*errp) {
- return;
- }
- }
-
/* write notes to vmcore */
write_elf32_notes(fd_write_vmcore, s, errp);
- if (*errp) {
- return;
- }
}
}
--
2.37.3

67
SOURCES/kvm-dump-Consolidate-elf-note-function.patch
Unescape View File

@ -0,0 +1,67 @@
From 0547599cf507930f91943f22d5f917ebacf69484 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:03 +0000
Subject: [PATCH 21/42] dump: Consolidate elf note function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [21/41] 52298c098c116aea75ad15894731ff412c2c4e73
Just like with the other write functions let's move the 32/64 bit elf
handling to a function to improve readability.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-10-frankja@linux.ibm.com>
(cherry picked from commit c68124738bc29017e4254c898bc40be7be477af7)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 88abde355a..a451abc590 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -520,6 +520,15 @@ static void write_elf_loads(DumpState *s, Error **errp)
}
}
+static void write_elf_notes(DumpState *s, Error **errp)
+{
+ if (dump_is_64bit(s)) {
+ write_elf64_notes(fd_write_vmcore, s, errp);
+ } else {
+ write_elf32_notes(fd_write_vmcore, s, errp);
+ }
+}
+
/* write elf header, PT_NOTE and elf note to vmcore. */
static void dump_begin(DumpState *s, Error **errp)
{
@@ -579,13 +588,8 @@ static void dump_begin(DumpState *s, Error **errp)
}
}
- if (dump_is_64bit(s)) {
- /* write notes to vmcore */
- write_elf64_notes(fd_write_vmcore, s, errp);
- } else {
- /* write notes to vmcore */
- write_elf32_notes(fd_write_vmcore, s, errp);
- }
+ /* write notes to vmcore */
+ write_elf_notes(s, errp);
}
static int get_next_block(DumpState *s, GuestPhysBlock *block)
--
2.37.3

169
SOURCES/kvm-dump-Consolidate-phdr-note-writes.patch
Unescape View File

@ -0,0 +1,169 @@
From f87abe1ef14e80731249ebe9fe1bea569a68e9b4 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:01 +0000
Subject: [PATCH 19/42] dump: Consolidate phdr note writes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [19/41] 180c4c0ab4941a0bf366dc7f32ee035e03daa6c0
There's no need to have two write functions. Let's rather have two
functions that set the data for elf 32/64 and then write it in a
common function.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-8-frankja@linux.ibm.com>
(cherry picked from commit bc7d558017e6700f9a05c61b0b638a8994945f0d)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 94 +++++++++++++++++++++++++++--------------------------
1 file changed, 48 insertions(+), 46 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 6394e94023..823ca32883 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -246,24 +246,15 @@ static void write_elf32_load(DumpState *s, MemoryMapping *memory_mapping,
}
}
-static void write_elf64_note(DumpState *s, Error **errp)
+static void write_elf64_phdr_note(DumpState *s, Elf64_Phdr *phdr)
{
- Elf64_Phdr phdr;
- int ret;
-
- memset(&phdr, 0, sizeof(Elf64_Phdr));
- phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump64(s, s->note_offset);
- phdr.p_paddr = 0;
- phdr.p_filesz = cpu_to_dump64(s, s->note_size);
- phdr.p_memsz = cpu_to_dump64(s, s->note_size);
- phdr.p_vaddr = 0;
-
- ret = fd_write_vmcore(&phdr, sizeof(Elf64_Phdr), s);
- if (ret < 0) {
- error_setg_errno(errp, -ret,
- "dump: failed to write program header table");
- }
+ memset(phdr, 0, sizeof(*phdr));
+ phdr->p_type = cpu_to_dump32(s, PT_NOTE);
+ phdr->p_offset = cpu_to_dump64(s, s->note_offset);
+ phdr->p_paddr = 0;
+ phdr->p_filesz = cpu_to_dump64(s, s->note_size);
+ phdr->p_memsz = cpu_to_dump64(s, s->note_size);
+ phdr->p_vaddr = 0;
}
static inline int cpu_index(CPUState *cpu)
@@ -311,24 +302,15 @@ static void write_elf64_notes(WriteCoreDumpFunction f, DumpState *s,
write_guest_note(f, s, errp);
}
-static void write_elf32_note(DumpState *s, Error **errp)
+static void write_elf32_phdr_note(DumpState *s, Elf32_Phdr *phdr)
{
- Elf32_Phdr phdr;
- int ret;
-
- memset(&phdr, 0, sizeof(Elf32_Phdr));
- phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump32(s, s->note_offset);
- phdr.p_paddr = 0;
- phdr.p_filesz = cpu_to_dump32(s, s->note_size);
- phdr.p_memsz = cpu_to_dump32(s, s->note_size);
- phdr.p_vaddr = 0;
-
- ret = fd_write_vmcore(&phdr, sizeof(Elf32_Phdr), s);
- if (ret < 0) {
- error_setg_errno(errp, -ret,
- "dump: failed to write program header table");
- }
+ memset(phdr, 0, sizeof(*phdr));
+ phdr->p_type = cpu_to_dump32(s, PT_NOTE);
+ phdr->p_offset = cpu_to_dump32(s, s->note_offset);
+ phdr->p_paddr = 0;
+ phdr->p_filesz = cpu_to_dump32(s, s->note_size);
+ phdr->p_memsz = cpu_to_dump32(s, s->note_size);
+ phdr->p_vaddr = 0;
}
static void write_elf32_notes(WriteCoreDumpFunction f, DumpState *s,
@@ -358,6 +340,32 @@ static void write_elf32_notes(WriteCoreDumpFunction f, DumpState *s,
write_guest_note(f, s, errp);
}
+static void write_elf_phdr_note(DumpState *s, Error **errp)
+{
+ ERRP_GUARD();
+ Elf32_Phdr phdr32;
+ Elf64_Phdr phdr64;
+ void *phdr;
+ size_t size;
+ int ret;
+
+ if (dump_is_64bit(s)) {
+ write_elf64_phdr_note(s, &phdr64);
+ size = sizeof(phdr64);
+ phdr = &phdr64;
+ } else {
+ write_elf32_phdr_note(s, &phdr32);
+ size = sizeof(phdr32);
+ phdr = &phdr32;
+ }
+
+ ret = fd_write_vmcore(phdr, size, s);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "dump: failed to write program header table");
+ }
+}
+
static void write_elf_section(DumpState *s, int type, Error **errp)
{
Elf32_Shdr shdr32;
@@ -551,13 +559,13 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- if (dump_is_64bit(s)) {
- /* write PT_NOTE to vmcore */
- write_elf64_note(s, errp);
- if (*errp) {
- return;
- }
+ /* write PT_NOTE to vmcore */
+ write_elf_phdr_note(s, errp);
+ if (*errp) {
+ return;
+ }
+ if (dump_is_64bit(s)) {
/* write all PT_LOAD to vmcore */
write_elf_loads(s, errp);
if (*errp) {
@@ -578,12 +586,6 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
} else {
- /* write PT_NOTE to vmcore */
- write_elf32_note(s, errp);
- if (*errp) {
- return;
- }
-
/* write all PT_LOAD to vmcore */
write_elf_loads(s, errp);
if (*errp) {
--
2.37.3

118
SOURCES/kvm-dump-Introduce-dump_is_64bit-helper-function.patch
Unescape View File

@ -0,0 +1,118 @@
From c851676d202b5b76962529f3b6d433936becbd8a Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:00 +0000
Subject: [PATCH 18/42] dump: Introduce dump_is_64bit() helper function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [18/41] a0fd2d1985c61b8e50d4a7ca26bc0ee6fcaa6196
Checking d_class in dump_info leads to lengthy conditionals so let's
shorten things a bit by introducing a helper function.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-7-frankja@linux.ibm.com>
(cherry picked from commit 05bbaa5040ccb3419e8b93af8040485430e2db42)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 85a402b38c..6394e94023 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -55,6 +55,11 @@ static Error *dump_migration_blocker;
DIV_ROUND_UP((name_size), 4) + \
DIV_ROUND_UP((desc_size), 4)) * 4)
+static inline bool dump_is_64bit(DumpState *s)
+{
+ return s->dump_info.d_class == ELFCLASS64;
+}
+
uint16_t cpu_to_dump16(DumpState *s, uint16_t val)
{
if (s->dump_info.d_endian == ELFDATA2LSB) {
@@ -489,7 +494,7 @@ static void write_elf_loads(DumpState *s, Error **errp)
get_offset_range(memory_mapping->phys_addr,
memory_mapping->length,
s, &offset, &filesz);
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
write_elf64_load(s, memory_mapping, phdr_index++, offset,
filesz, errp);
} else {
@@ -537,7 +542,7 @@ static void dump_begin(DumpState *s, Error **errp)
*/
/* write elf header to vmcore */
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
write_elf64_header(s, errp);
} else {
write_elf32_header(s, errp);
@@ -546,7 +551,7 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
/* write PT_NOTE to vmcore */
write_elf64_note(s, errp);
if (*errp) {
@@ -757,7 +762,7 @@ static void get_note_sizes(DumpState *s, const void *note,
uint64_t name_sz;
uint64_t desc_sz;
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
const Elf64_Nhdr *hdr = note;
note_head_sz = sizeof(Elf64_Nhdr);
name_sz = tswap64(hdr->n_namesz);
@@ -1017,10 +1022,10 @@ out:
static void write_dump_header(DumpState *s, Error **errp)
{
- if (s->dump_info.d_class == ELFCLASS32) {
- create_header32(s, errp);
- } else {
+ if (dump_is_64bit(s)) {
create_header64(s, errp);
+ } else {
+ create_header32(s, errp);
}
}
@@ -1715,8 +1720,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
uint32_t size;
uint16_t format;
- note_head_size = s->dump_info.d_class == ELFCLASS32 ?
- sizeof(Elf32_Nhdr) : sizeof(Elf64_Nhdr);
+ note_head_size = dump_is_64bit(s) ?
+ sizeof(Elf64_Nhdr) : sizeof(Elf32_Nhdr);
format = le16_to_cpu(vmci->vmcoreinfo.guest_format);
size = le32_to_cpu(vmci->vmcoreinfo.size);
@@ -1819,7 +1824,7 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
}
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
s->phdr_offset = sizeof(Elf64_Ehdr);
s->shdr_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
s->note_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
--
2.37.3

136
SOURCES/kvm-dump-Introduce-shdr_num-to-decrease-complexity.patch
Unescape View File

@ -0,0 +1,136 @@
From 255722667a4fa4d522bb0b7e0825cbbe635abb8d Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:57 +0000
Subject: [PATCH 15/42] dump: Introduce shdr_num to decrease complexity
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [15/41] b0215ea5d381ef7f6abfe3f3bafea51ce933da56
Let's move from a boolean to a int variable which will later enable us
to store the number of sections that are in the dump file.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-4-frankja@linux.ibm.com>
(cherry picked from commit 862a395858e5a302ed5921487777acdc95a3a31b)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 24 ++++++++++++------------
include/sysemu/dump.h | 2 +-
2 files changed, 13 insertions(+), 13 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 7236b167cc..972e28b089 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -145,12 +145,12 @@ static void write_elf64_header(DumpState *s, Error **errp)
elf_header.e_phoff = cpu_to_dump64(s, sizeof(Elf64_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
- if (s->have_section) {
+ if (s->shdr_num) {
uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump64(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, 1);
+ elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
@@ -182,12 +182,12 @@ static void write_elf32_header(DumpState *s, Error **errp)
elf_header.e_phoff = cpu_to_dump32(s, sizeof(Elf32_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
- if (s->have_section) {
+ if (s->shdr_num) {
uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump32(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, 1);
+ elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
@@ -566,7 +566,7 @@ static void dump_begin(DumpState *s, Error **errp)
}
/* write section to vmcore */
- if (s->have_section) {
+ if (s->shdr_num) {
write_elf_section(s, 1, errp);
if (*errp) {
return;
@@ -592,7 +592,7 @@ static void dump_begin(DumpState *s, Error **errp)
}
/* write section to vmcore */
- if (s->have_section) {
+ if (s->shdr_num) {
write_elf_section(s, 0, errp);
if (*errp) {
return;
@@ -1811,11 +1811,11 @@ static void dump_init(DumpState *s, int fd, bool has_format,
*/
s->phdr_num = 1; /* PT_NOTE */
if (s->list.num < UINT16_MAX - 2) {
+ s->shdr_num = 0;
s->phdr_num += s->list.num;
- s->have_section = false;
} else {
/* sh_info of section 0 holds the real number of phdrs */
- s->have_section = true;
+ s->shdr_num = 1;
/* the type of shdr->sh_info is uint32_t, so we should avoid overflow */
if (s->list.num <= UINT32_MAX - 1) {
@@ -1826,19 +1826,19 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (s->dump_info.d_class == ELFCLASS64) {
- if (s->have_section) {
+ if (s->shdr_num) {
s->memory_offset = sizeof(Elf64_Ehdr) +
sizeof(Elf64_Phdr) * s->phdr_num +
- sizeof(Elf64_Shdr) + s->note_size;
+ sizeof(Elf64_Shdr) * s->shdr_num + s->note_size;
} else {
s->memory_offset = sizeof(Elf64_Ehdr) +
sizeof(Elf64_Phdr) * s->phdr_num + s->note_size;
}
} else {
- if (s->have_section) {
+ if (s->shdr_num) {
s->memory_offset = sizeof(Elf32_Ehdr) +
sizeof(Elf32_Phdr) * s->phdr_num +
- sizeof(Elf32_Shdr) + s->note_size;
+ sizeof(Elf32_Shdr) * s->shdr_num + s->note_size;
} else {
s->memory_offset = sizeof(Elf32_Ehdr) +
sizeof(Elf32_Phdr) * s->phdr_num + s->note_size;
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index b463fc9c02..19458bffbd 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -155,7 +155,7 @@ typedef struct DumpState {
ArchDumpInfo dump_info;
MemoryMappingList list;
uint32_t phdr_num;
- bool have_section;
+ uint32_t shdr_num;
bool resume;
bool detached;
ssize_t note_size;
--
2.37.3

142
SOURCES/kvm-dump-Refactor-dump_iterate-and-introduce-dump_filter.patch
Unescape View File

@ -0,0 +1,142 @@
From a18ba2fbaf132724e81be92da42b36d8f365e66c Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:56 +0000
Subject: [PATCH 24/42] dump: Refactor dump_iterate and introduce
dump_filter_memblock_*()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [24/41] 74ef470f24d9d98093c4d63730a99474587033fd
The iteration over the memblocks in dump_iterate() is hard to
understand so it's about time to clean it up. Instead of manually
grabbing the next memblock we can use QTAILQ_FOREACH to iterate over
all memblocks.
Additionally we move the calculation of the offset and length out by
introducing and using the dump_filter_memblock_*() functions. These
functions will later be used to cleanup other parts of dump.c.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-4-frankja@linux.ibm.com>
(cherry picked from commit 1e8113032f5b1efc5da66382470ce4809c76f8f2)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 74 ++++++++++++++++++++++++++++++-----------------------
1 file changed, 42 insertions(+), 32 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index fa787f379f..d981e843dd 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -592,31 +592,43 @@ static void dump_begin(DumpState *s, Error **errp)
write_elf_notes(s, errp);
}
-static int get_next_block(DumpState *s, GuestPhysBlock *block)
+static int64_t dump_filtered_memblock_size(GuestPhysBlock *block,
+ int64_t filter_area_start,
+ int64_t filter_area_length)
{
- while (1) {
- block = QTAILQ_NEXT(block, next);
- if (!block) {
- /* no more block */
- return 1;
- }
+ int64_t size, left, right;
- s->start = 0;
- s->next_block = block;
- if (s->has_filter) {
- if (block->target_start >= s->begin + s->length ||
- block->target_end <= s->begin) {
- /* This block is out of the range */
- continue;
- }
+ /* No filter, return full size */
+ if (!filter_area_length) {
+ return block->target_end - block->target_start;
+ }
- if (s->begin > block->target_start) {
- s->start = s->begin - block->target_start;
- }
+ /* calculate the overlapped region. */
+ left = MAX(filter_area_start, block->target_start);
+ right = MIN(filter_area_start + filter_area_length, block->target_end);
+ size = right - left;
+ size = size > 0 ? size : 0;
+
+ return size;
+}
+
+static int64_t dump_filtered_memblock_start(GuestPhysBlock *block,
+ int64_t filter_area_start,
+ int64_t filter_area_length)
+{
+ if (filter_area_length) {
+ /* return -1 if the block is not within filter area */
+ if (block->target_start >= filter_area_start + filter_area_length ||
+ block->target_end <= filter_area_start) {
+ return -1;
}
- return 0;
+ if (filter_area_start > block->target_start) {
+ return filter_area_start - block->target_start;
+ }
}
+
+ return 0;
}
/* write all memory to vmcore */
@@ -624,24 +636,22 @@ static void dump_iterate(DumpState *s, Error **errp)
{
ERRP_GUARD();
GuestPhysBlock *block;
- int64_t size;
-
- do {
- block = s->next_block;
+ int64_t memblock_size, memblock_start;
- size = block->target_end - block->target_start;
- if (s->has_filter) {
- size -= s->start;
- if (s->begin + s->length < block->target_end) {
- size -= block->target_end - (s->begin + s->length);
- }
+ QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
+ memblock_start = dump_filtered_memblock_start(block, s->begin, s->length);
+ if (memblock_start == -1) {
+ continue;
}
- write_memory(s, block, s->start, size, errp);
+
+ memblock_size = dump_filtered_memblock_size(block, s->begin, s->length);
+
+ /* Write the memory to file */
+ write_memory(s, block, memblock_start, memblock_size, errp);
if (*errp) {
return;
}
-
- } while (!get_next_block(s, block));
+ }
}
static void create_vmcore(DumpState *s, Error **errp)
--
2.37.3

45
SOURCES/kvm-dump-Reintroduce-memory_offset-and-section_offset.patch
Unescape View File

@ -0,0 +1,45 @@
From 6932fe3afbec443bbf6acff5b707536254e1bc37 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:16 +0000
Subject: [PATCH 35/42] dump: Reintroduce memory_offset and section_offset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [35/41] e60c0d066aeeedb42e724712bc3aa7b7591c6c79
section_offset will later be used to store the offset to the section
data which will be stored last. For now memory_offset is only needed
to make section_offset look nicer.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-5-frankja@linux.ibm.com>
(cherry picked from commit 13fd417ddc81a1685c6a8f4e1c80bbfe7150f164)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/dump/dump.c b/dump/dump.c
index d17537d4e9..7a42401790 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1885,6 +1885,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
s->phdr_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
s->note_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
}
+ s->memory_offset = s->note_offset + s->note_size;
+ s->section_offset = s->memory_offset + s->total_size;
return;
--
2.37.3

70
SOURCES/kvm-dump-Remove-the-section-if-when-calculating-the-memo.patch
Unescape View File

@ -0,0 +1,70 @@
From a8eeab6936a2bd27b33b63aed7e2ef96034f7772 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:58 +0000
Subject: [PATCH 16/42] dump: Remove the section if when calculating the memory
offset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [16/41] ff214d2c23b9cb16fd49d22d976829267df43133
When s->shdr_num is 0 we'll add 0 bytes of section headers which is
equivalent to not adding section headers but with the multiplication
we can remove a if/else.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-5-frankja@linux.ibm.com>
(cherry picked from commit 344107e07bd81546474a54ab83800158ca953059)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 24 ++++++++----------------
1 file changed, 8 insertions(+), 16 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 972e28b089..5cc2322325 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1826,23 +1826,15 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (s->dump_info.d_class == ELFCLASS64) {
- if (s->shdr_num) {
- s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->phdr_num +
- sizeof(Elf64_Shdr) * s->shdr_num + s->note_size;
- } else {
- s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->phdr_num + s->note_size;
- }
+ s->memory_offset = sizeof(Elf64_Ehdr) +
+ sizeof(Elf64_Phdr) * s->phdr_num +
+ sizeof(Elf64_Shdr) * s->shdr_num +
+ s->note_size;
} else {
- if (s->shdr_num) {
- s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->phdr_num +
- sizeof(Elf32_Shdr) * s->shdr_num + s->note_size;
- } else {
- s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->phdr_num + s->note_size;
- }
+ s->memory_offset = sizeof(Elf32_Ehdr) +
+ sizeof(Elf32_Phdr) * s->phdr_num +
+ sizeof(Elf32_Shdr) * s->shdr_num +
+ s->note_size;
}
return;
--
2.37.3

176
SOURCES/kvm-dump-Remove-the-sh_info-variable.patch
Unescape View File

@ -0,0 +1,176 @@
From eb763bec53d6b9aea7a6b60b0cf8c5d8b5f1b35c Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 7 Apr 2022 09:48:24 +0000
Subject: [PATCH 14/42] dump: Remove the sh_info variable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [14/41] 24af12b78c8f5a02cf85df2f6b1d64249f9499c9
There's no need to have phdr_num and sh_info at the same time. We can
make phdr_num 32 bit and set PN_XNUM when we write the header if
phdr_num >= PN_XNUM.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220407094824.5074-1-frankja@linux.ibm.com>
(cherry picked from commit 046bc4160bc780eaacc2d702a2589f1a7a01188d)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 44 +++++++++++++++++++++++--------------------
include/sysemu/dump.h | 3 +--
2 files changed, 25 insertions(+), 22 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 9876123f2e..7236b167cc 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -124,6 +124,12 @@ static int fd_write_vmcore(const void *buf, size_t size, void *opaque)
static void write_elf64_header(DumpState *s, Error **errp)
{
+ /*
+ * phnum in the elf header is 16 bit, if we have more segments we
+ * set phnum to PN_XNUM and write the real number of segments to a
+ * special section.
+ */
+ uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
Elf64_Ehdr elf_header;
int ret;
@@ -138,9 +144,9 @@ static void write_elf64_header(DumpState *s, Error **errp)
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
elf_header.e_phoff = cpu_to_dump64(s, sizeof(Elf64_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, s->phdr_num);
+ elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->have_section) {
- uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->sh_info;
+ uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump64(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
@@ -155,6 +161,12 @@ static void write_elf64_header(DumpState *s, Error **errp)
static void write_elf32_header(DumpState *s, Error **errp)
{
+ /*
+ * phnum in the elf header is 16 bit, if we have more segments we
+ * set phnum to PN_XNUM and write the real number of segments to a
+ * special section.
+ */
+ uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
Elf32_Ehdr elf_header;
int ret;
@@ -169,9 +181,9 @@ static void write_elf32_header(DumpState *s, Error **errp)
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
elf_header.e_phoff = cpu_to_dump32(s, sizeof(Elf32_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, s->phdr_num);
+ elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->have_section) {
- uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->sh_info;
+ uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump32(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
@@ -358,12 +370,12 @@ static void write_elf_section(DumpState *s, int type, Error **errp)
if (type == 0) {
shdr_size = sizeof(Elf32_Shdr);
memset(&shdr32, 0, shdr_size);
- shdr32.sh_info = cpu_to_dump32(s, s->sh_info);
+ shdr32.sh_info = cpu_to_dump32(s, s->phdr_num);
shdr = &shdr32;
} else {
shdr_size = sizeof(Elf64_Shdr);
memset(&shdr64, 0, shdr_size);
- shdr64.sh_info = cpu_to_dump32(s, s->sh_info);
+ shdr64.sh_info = cpu_to_dump32(s, s->phdr_num);
shdr = &shdr64;
}
@@ -478,13 +490,6 @@ static void write_elf_loads(DumpState *s, Error **errp)
hwaddr offset, filesz;
MemoryMapping *memory_mapping;
uint32_t phdr_index = 1;
- uint32_t max_index;
-
- if (s->have_section) {
- max_index = s->sh_info;
- } else {
- max_index = s->phdr_num;
- }
QTAILQ_FOREACH(memory_mapping, &s->list.head, next) {
get_offset_range(memory_mapping->phys_addr,
@@ -502,7 +507,7 @@ static void write_elf_loads(DumpState *s, Error **errp)
return;
}
- if (phdr_index >= max_index) {
+ if (phdr_index >= s->phdr_num) {
break;
}
}
@@ -1809,22 +1814,21 @@ static void dump_init(DumpState *s, int fd, bool has_format,
s->phdr_num += s->list.num;
s->have_section = false;
} else {
+ /* sh_info of section 0 holds the real number of phdrs */
s->have_section = true;
- s->phdr_num = PN_XNUM;
- s->sh_info = 1; /* PT_NOTE */
/* the type of shdr->sh_info is uint32_t, so we should avoid overflow */
if (s->list.num <= UINT32_MAX - 1) {
- s->sh_info += s->list.num;
+ s->phdr_num += s->list.num;
} else {
- s->sh_info = UINT32_MAX;
+ s->phdr_num = UINT32_MAX;
}
}
if (s->dump_info.d_class == ELFCLASS64) {
if (s->have_section) {
s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->sh_info +
+ sizeof(Elf64_Phdr) * s->phdr_num +
sizeof(Elf64_Shdr) + s->note_size;
} else {
s->memory_offset = sizeof(Elf64_Ehdr) +
@@ -1833,7 +1837,7 @@ static void dump_init(DumpState *s, int fd, bool has_format,
} else {
if (s->have_section) {
s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->sh_info +
+ sizeof(Elf32_Phdr) * s->phdr_num +
sizeof(Elf32_Shdr) + s->note_size;
} else {
s->memory_offset = sizeof(Elf32_Ehdr) +
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 250143cb5a..b463fc9c02 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -154,8 +154,7 @@ typedef struct DumpState {
GuestPhysBlockList guest_phys_blocks;
ArchDumpInfo dump_info;
MemoryMappingList list;
- uint16_t phdr_num;
- uint32_t sh_info;
+ uint32_t phdr_num;
bool have_section;
bool resume;
bool detached;
--
2.37.3

69
SOURCES/kvm-dump-Rename-write_elf-_phdr_note-to-prepare_elf-_phd.patch
Unescape View File

@ -0,0 +1,69 @@
From 18fef7f02801d51207d67b8f8ec5f0d828889c78 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:11:01 +0000
Subject: [PATCH 29/42] dump: Rename write_elf*_phdr_note to
prepare_elf*_phdr_note
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [29/41] 876cea6f6e51be8df2763f56d0daef99d11fdd49
The functions in question do not actually write to the file descriptor
they set up a buffer which is later written to the fd.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-9-frankja@linux.ibm.com>
(cherry picked from commit 2341a94d3a0a8a93a5a977e642da1807b8edaab8)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 8d5226f861..c2c1341ad7 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -261,7 +261,7 @@ static void write_elf32_load(DumpState *s, MemoryMapping *memory_mapping,
}
}
-static void write_elf64_phdr_note(DumpState *s, Elf64_Phdr *phdr)
+static void prepare_elf64_phdr_note(DumpState *s, Elf64_Phdr *phdr)
{
memset(phdr, 0, sizeof(*phdr));
phdr->p_type = cpu_to_dump32(s, PT_NOTE);
@@ -317,7 +317,7 @@ static void write_elf64_notes(WriteCoreDumpFunction f, DumpState *s,
write_guest_note(f, s, errp);
}
-static void write_elf32_phdr_note(DumpState *s, Elf32_Phdr *phdr)
+static void prepare_elf32_phdr_note(DumpState *s, Elf32_Phdr *phdr)
{
memset(phdr, 0, sizeof(*phdr));
phdr->p_type = cpu_to_dump32(s, PT_NOTE);
@@ -365,11 +365,11 @@ static void write_elf_phdr_note(DumpState *s, Error **errp)
int ret;
if (dump_is_64bit(s)) {
- write_elf64_phdr_note(s, &phdr64);
+ prepare_elf64_phdr_note(s, &phdr64);
size = sizeof(phdr64);
phdr = &phdr64;
} else {
- write_elf32_phdr_note(s, &phdr32);
+ prepare_elf32_phdr_note(s, &phdr32);
size = sizeof(phdr32);
phdr = &phdr32;
}
--
2.37.3

57
SOURCES/kvm-dump-Rename-write_elf_loads-to-write_elf_phdr_loads.patch
Unescape View File

@ -0,0 +1,57 @@
From 04d4947a22fe3192384ff486d0a979d799ded98e Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:55 +0000
Subject: [PATCH 23/42] dump: Rename write_elf_loads to write_elf_phdr_loads
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [23/41] 18e3ef70b97c525b7c43cf12143204bdb1060e4f
Let's make it a bit clearer that we write the program headers of the
PT_LOAD type.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@ibm.linux.com>
Message-Id: <20220811121111.9878-3-frankja@linux.ibm.com>
(cherry picked from commit afae6056ea79e2d89fd90867de3a01732eae724f)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index a451abc590..fa787f379f 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -491,7 +491,7 @@ static void get_offset_range(hwaddr phys_addr,
}
}
-static void write_elf_loads(DumpState *s, Error **errp)
+static void write_elf_phdr_loads(DumpState *s, Error **errp)
{
ERRP_GUARD();
hwaddr offset, filesz;
@@ -574,8 +574,8 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- /* write all PT_LOAD to vmcore */
- write_elf_loads(s, errp);
+ /* write all PT_LOADs to vmcore */
+ write_elf_phdr_loads(s, errp);
if (*errp) {
return;
}
--
2.37.3

68
SOURCES/kvm-dump-Reorder-struct-DumpState.patch
Unescape View File

@ -0,0 +1,68 @@
From 7e8d6290099b33f88621b45e62652a97704c9573 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:15 +0000
Subject: [PATCH 34/42] dump: Reorder struct DumpState
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [34/41] 8d44e5e8c86ea5b33644eba141046cd657d0071e
Let's move ELF related members into one block and guest memory related
ones into another to improve readability.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-4-frankja@linux.ibm.com>
(cherry picked from commit 8384b73c46fd474847d7e74d121318e344edc3c4)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/sysemu/dump.h | 16 +++++++++-------
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 9995f65dc8..9ed811b313 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -154,15 +154,8 @@ typedef struct DumpState {
GuestPhysBlockList guest_phys_blocks;
ArchDumpInfo dump_info;
MemoryMappingList list;
- uint32_t phdr_num;
- uint32_t shdr_num;
bool resume;
bool detached;
- ssize_t note_size;
- hwaddr shdr_offset;
- hwaddr phdr_offset;
- hwaddr section_offset;
- hwaddr note_offset;
hwaddr memory_offset;
int fd;
@@ -177,6 +170,15 @@ typedef struct DumpState {
int64_t filter_area_begin; /* Start address of partial guest memory area */
int64_t filter_area_length; /* Length of partial guest memory area */
+ /* Elf dump related data */
+ uint32_t phdr_num;
+ uint32_t shdr_num;
+ ssize_t note_size;
+ hwaddr shdr_offset;
+ hwaddr phdr_offset;
+ hwaddr section_offset;
+ hwaddr note_offset;
+
void *elf_section_hdrs; /* Pointer to section header buffer */
uint8_t *note_buf; /* buffer for notes */
--
2.37.3

467
SOURCES/kvm-dump-Replace-opaque-DumpState-pointer-with-a-typed-o.patch
Unescape View File

@ -0,0 +1,467 @@
From 8f674e0e12e4b88fc035948612a0b0949e0ad892 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:54 +0000
Subject: [PATCH 22/42] dump: Replace opaque DumpState pointer with a typed one
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [22/41] 5f071d7ef441ae6f5da70eb56018c4657deee3d7
It's always better to convey the type of a pointer if at all
possible. So let's add the DumpState typedef to typedefs.h and move
the dump note functions from the opaque pointers to DumpState
pointers.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
CC: Peter Maydell <peter.maydell@linaro.org>
CC: Cédric Le Goater <clg@kaod.org>
CC: Daniel Henrique Barboza <danielhb413@gmail.com>
CC: David Gibson <david@gibson.dropbear.id.au>
CC: Greg Kurz <groug@kaod.org>
CC: Palmer Dabbelt <palmer@dabbelt.com>
CC: Alistair Francis <alistair.francis@wdc.com>
CC: Bin Meng <bin.meng@windriver.com>
CC: Cornelia Huck <cohuck@redhat.com>
CC: Thomas Huth <thuth@redhat.com>
CC: Richard Henderson <richard.henderson@linaro.org>
CC: David Hildenbrand <david@redhat.com>
Acked-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-2-frankja@linux.ibm.com>
(cherry picked from commit 1af0006ab959864dfa2f59e9136c5fb93000b61f)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/hw/core/sysemu-cpu-ops.h | 8 ++++----
include/qemu/typedefs.h | 1 +
target/arm/arch_dump.c | 6 ++----
target/arm/cpu.h | 4 ++--
target/i386/arch_dump.c | 30 +++++++++++++++---------------
target/i386/cpu.h | 8 ++++----
target/ppc/arch_dump.c | 18 +++++++++---------
target/ppc/cpu.h | 4 ++--
target/riscv/arch_dump.c | 6 ++----
target/riscv/cpu.h | 4 ++--
target/s390x/arch_dump.c | 10 +++++-----
target/s390x/s390x-internal.h | 2 +-
12 files changed, 49 insertions(+), 52 deletions(-)
diff --git a/include/hw/core/sysemu-cpu-ops.h b/include/hw/core/sysemu-cpu-ops.h
index a9ba39e5f2..ee169b872c 100644
--- a/include/hw/core/sysemu-cpu-ops.h
+++ b/include/hw/core/sysemu-cpu-ops.h
@@ -53,25 +53,25 @@ typedef struct SysemuCPUOps {
* 32-bit VM coredump.
*/
int (*write_elf32_note)(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
/**
* @write_elf64_note: Callback for writing a CPU-specific ELF note to a
* 64-bit VM coredump.
*/
int (*write_elf64_note)(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
/**
* @write_elf32_qemunote: Callback for writing a CPU- and QEMU-specific ELF
* note to a 32-bit VM coredump.
*/
int (*write_elf32_qemunote)(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
/**
* @write_elf64_qemunote: Callback for writing a CPU- and QEMU-specific ELF
* note to a 64-bit VM coredump.
*/
int (*write_elf64_qemunote)(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
/**
* @virtio_is_big_endian: Callback to return %true if a CPU which supports
* runtime configurable endianness is currently big-endian.
diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h
index ee60eb3de4..ac9d031be6 100644
--- a/include/qemu/typedefs.h
+++ b/include/qemu/typedefs.h
@@ -125,6 +125,7 @@ typedef struct VirtIODevice VirtIODevice;
typedef struct Visitor Visitor;
typedef struct VMChangeStateEntry VMChangeStateEntry;
typedef struct VMStateDescription VMStateDescription;
+typedef struct DumpState DumpState;
/*
* Pointer types
diff --git a/target/arm/arch_dump.c b/target/arm/arch_dump.c
index 0184845310..3a824e0aa6 100644
--- a/target/arm/arch_dump.c
+++ b/target/arm/arch_dump.c
@@ -232,12 +232,11 @@ static int aarch64_write_elf64_sve(WriteCoreDumpFunction f,
#endif
int arm_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct aarch64_note note;
ARMCPU *cpu = ARM_CPU(cs);
CPUARMState *env = &cpu->env;
- DumpState *s = opaque;
uint64_t pstate, sp;
int ret, i;
@@ -360,12 +359,11 @@ static int arm_write_elf32_vfp(WriteCoreDumpFunction f, CPUARMState *env,
}
int arm_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct arm_note note;
ARMCPU *cpu = ARM_CPU(cs);
CPUARMState *env = &cpu->env;
- DumpState *s = opaque;
int ret, i;
bool fpvalid = cpu_isar_feature(aa32_vfp_simd, cpu);
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index e33f37b70a..8d2f496ef9 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -1065,9 +1065,9 @@ int arm_gen_dynamic_svereg_xml(CPUState *cpu, int base_reg);
const char *arm_gdb_get_dynamic_xml(CPUState *cpu, const char *xmlname);
int arm_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int arm_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
#ifdef TARGET_AARCH64
int aarch64_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
diff --git a/target/i386/arch_dump.c b/target/i386/arch_dump.c
index 004141fc04..c290910a04 100644
--- a/target/i386/arch_dump.c
+++ b/target/i386/arch_dump.c
@@ -42,7 +42,7 @@ typedef struct {
static int x86_64_write_elf64_note(WriteCoreDumpFunction f,
CPUX86State *env, int id,
- void *opaque)
+ DumpState *s)
{
x86_64_user_regs_struct regs;
Elf64_Nhdr *note;
@@ -94,7 +94,7 @@ static int x86_64_write_elf64_note(WriteCoreDumpFunction f,
buf += descsz - sizeof(x86_64_user_regs_struct)-sizeof(target_ulong);
memcpy(buf, &regs, sizeof(x86_64_user_regs_struct));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -148,7 +148,7 @@ static void x86_fill_elf_prstatus(x86_elf_prstatus *prstatus, CPUX86State *env,
}
static int x86_write_elf64_note(WriteCoreDumpFunction f, CPUX86State *env,
- int id, void *opaque)
+ int id, DumpState *s)
{
x86_elf_prstatus prstatus;
Elf64_Nhdr *note;
@@ -170,7 +170,7 @@ static int x86_write_elf64_note(WriteCoreDumpFunction f, CPUX86State *env,
buf += ROUND_UP(name_size, 4);
memcpy(buf, &prstatus, sizeof(prstatus));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -180,7 +180,7 @@ static int x86_write_elf64_note(WriteCoreDumpFunction f, CPUX86State *env,
}
int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
int ret;
@@ -189,10 +189,10 @@ int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
bool lma = !!(first_x86_cpu->env.hflags & HF_LMA_MASK);
if (lma) {
- ret = x86_64_write_elf64_note(f, &cpu->env, cpuid, opaque);
+ ret = x86_64_write_elf64_note(f, &cpu->env, cpuid, s);
} else {
#endif
- ret = x86_write_elf64_note(f, &cpu->env, cpuid, opaque);
+ ret = x86_write_elf64_note(f, &cpu->env, cpuid, s);
#ifdef TARGET_X86_64
}
#endif
@@ -201,7 +201,7 @@ int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
}
int x86_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
x86_elf_prstatus prstatus;
@@ -224,7 +224,7 @@ int x86_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
buf += ROUND_UP(name_size, 4);
memcpy(buf, &prstatus, sizeof(prstatus));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -329,7 +329,7 @@ static void qemu_get_cpustate(QEMUCPUState *s, CPUX86State *env)
static inline int cpu_write_qemu_note(WriteCoreDumpFunction f,
CPUX86State *env,
- void *opaque,
+ DumpState *s,
int type)
{
QEMUCPUState state;
@@ -369,7 +369,7 @@ static inline int cpu_write_qemu_note(WriteCoreDumpFunction f,
buf += ROUND_UP(name_size, 4);
memcpy(buf, &state, sizeof(state));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -379,19 +379,19 @@ static inline int cpu_write_qemu_note(WriteCoreDumpFunction f,
}
int x86_cpu_write_elf64_qemunote(WriteCoreDumpFunction f, CPUState *cs,
- void *opaque)
+ DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
- return cpu_write_qemu_note(f, &cpu->env, opaque, 1);
+ return cpu_write_qemu_note(f, &cpu->env, s, 1);
}
int x86_cpu_write_elf32_qemunote(WriteCoreDumpFunction f, CPUState *cs,
- void *opaque)
+ DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
- return cpu_write_qemu_note(f, &cpu->env, opaque, 0);
+ return cpu_write_qemu_note(f, &cpu->env, s, 0);
}
int cpu_get_dump_info(ArchDumpInfo *info,
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 006b735fe4..5d2ddd81b9 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1887,13 +1887,13 @@ extern const VMStateDescription vmstate_x86_cpu;
int x86_cpu_pending_interrupt(CPUState *cs, int interrupt_request);
int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int x86_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int x86_cpu_write_elf64_qemunote(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
int x86_cpu_write_elf32_qemunote(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
void x86_cpu_get_memory_mapping(CPUState *cpu, MemoryMappingList *list,
Error **errp);
diff --git a/target/ppc/arch_dump.c b/target/ppc/arch_dump.c
index bb392f6d88..e9f512bcd4 100644
--- a/target/ppc/arch_dump.c
+++ b/target/ppc/arch_dump.c
@@ -270,23 +270,23 @@ ssize_t cpu_get_note_size(int class, int machine, int nr_cpus)
static int ppc_write_all_elf_notes(const char *note_name,
WriteCoreDumpFunction f,
PowerPCCPU *cpu, int id,
- void *opaque)
+ DumpState *s)
{
- NoteFuncArg arg = { .state = opaque };
+ NoteFuncArg arg = { .state = s };
int ret = -1;
int note_size;
const NoteFuncDesc *nf;
for (nf = note_func; nf->note_contents_func; nf++) {
- arg.note.hdr.n_namesz = cpu_to_dump32(opaque, sizeof(arg.note.name));
- arg.note.hdr.n_descsz = cpu_to_dump32(opaque, nf->contents_size);
+ arg.note.hdr.n_namesz = cpu_to_dump32(s, sizeof(arg.note.name));
+ arg.note.hdr.n_descsz = cpu_to_dump32(s, nf->contents_size);
strncpy(arg.note.name, note_name, sizeof(arg.note.name));
(*nf->note_contents_func)(&arg, cpu);
note_size =
sizeof(arg.note) - sizeof(arg.note.contents) + nf->contents_size;
- ret = f(&arg.note, note_size, opaque);
+ ret = f(&arg.note, note_size, s);
if (ret < 0) {
return -1;
}
@@ -295,15 +295,15 @@ static int ppc_write_all_elf_notes(const char *note_name,
}
int ppc64_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
- return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, opaque);
+ return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, s);
}
int ppc32_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
- return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, opaque);
+ return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, s);
}
diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
index 23e8b76c85..f5fb284706 100644
--- a/target/ppc/cpu.h
+++ b/target/ppc/cpu.h
@@ -1289,9 +1289,9 @@ void ppc_gdb_gen_spr_xml(PowerPCCPU *cpu);
const char *ppc_gdb_get_dynamic_xml(CPUState *cs, const char *xml_name);
#endif
int ppc64_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int ppc32_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
#ifndef CONFIG_USER_ONLY
void ppc_cpu_do_interrupt(CPUState *cpu);
bool ppc_cpu_exec_interrupt(CPUState *cpu, int int_req);
diff --git a/target/riscv/arch_dump.c b/target/riscv/arch_dump.c
index 709f621d82..736a232956 100644
--- a/target/riscv/arch_dump.c
+++ b/target/riscv/arch_dump.c
@@ -64,12 +64,11 @@ static void riscv64_note_init(struct riscv64_note *note, DumpState *s,
}
int riscv_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct riscv64_note note;
RISCVCPU *cpu = RISCV_CPU(cs);
CPURISCVState *env = &cpu->env;
- DumpState *s = opaque;
int ret, i = 0;
const char name[] = "CORE";
@@ -134,12 +133,11 @@ static void riscv32_note_init(struct riscv32_note *note, DumpState *s,
}
int riscv_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct riscv32_note note;
RISCVCPU *cpu = RISCV_CPU(cs);
CPURISCVState *env = &cpu->env;
- DumpState *s = opaque;
int ret, i;
const char name[] = "CORE";
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index 0760c0af93..4cce524b2c 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -344,9 +344,9 @@ extern const char * const riscv_fpr_regnames[];
const char *riscv_cpu_get_trap_name(target_ulong cause, bool async);
void riscv_cpu_do_interrupt(CPUState *cpu);
int riscv_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int riscv_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int riscv_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
int riscv_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
bool riscv_cpu_fp_enabled(CPURISCVState *env);
diff --git a/target/s390x/arch_dump.c b/target/s390x/arch_dump.c
index 08daf93ae1..f60a14920d 100644
--- a/target/s390x/arch_dump.c
+++ b/target/s390x/arch_dump.c
@@ -204,7 +204,7 @@ static const NoteFuncDesc note_linux[] = {
static int s390x_write_elf64_notes(const char *note_name,
WriteCoreDumpFunction f,
S390CPU *cpu, int id,
- void *opaque,
+ DumpState *s,
const NoteFuncDesc *funcs)
{
Note note;
@@ -222,7 +222,7 @@ static int s390x_write_elf64_notes(const char *note_name,
(*nf->note_contents_func)(&note, cpu, id);
note_size = sizeof(note) - sizeof(note.contents) + nf->contents_size;
- ret = f(&note, note_size, opaque);
+ ret = f(&note, note_size, s);
if (ret < 0) {
return -1;
@@ -235,16 +235,16 @@ static int s390x_write_elf64_notes(const char *note_name,
int s390_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
S390CPU *cpu = S390_CPU(cs);
int r;
- r = s390x_write_elf64_notes("CORE", f, cpu, cpuid, opaque, note_core);
+ r = s390x_write_elf64_notes("CORE", f, cpu, cpuid, s, note_core);
if (r) {
return r;
}
- return s390x_write_elf64_notes("LINUX", f, cpu, cpuid, opaque, note_linux);
+ return s390x_write_elf64_notes("LINUX", f, cpu, cpuid, s, note_linux);
}
int cpu_get_dump_info(ArchDumpInfo *info,
diff --git a/target/s390x/s390x-internal.h b/target/s390x/s390x-internal.h
index 1a178aed41..02cf6c3f43 100644
--- a/target/s390x/s390x-internal.h
+++ b/target/s390x/s390x-internal.h
@@ -228,7 +228,7 @@ static inline hwaddr decode_basedisp_s(CPUS390XState *env, uint32_t ipb,
/* arch_dump.c */
int s390_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
/* cc_helper.c */
--
2.37.3

73
SOURCES/kvm-dump-Rework-dump_calculate_size-function.patch
Unescape View File

@ -0,0 +1,73 @@
From 1f7cb73592a1922b3a981eb3232098281e07679f Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:59 +0000
Subject: [PATCH 27/42] dump: Rework dump_calculate_size function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [27/41] eaa05c39109b57a119752ad3df66f4c2ace2cbe4
dump_calculate_size() sums up all the sizes of the guest memory
blocks. Since we already have a function that calculates the size of a
single memory block (dump_get_memblock_size()) we can simply iterate
over the blocks and use the function instead of calculating the size
ourselves.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Message-Id: <20220811121111.9878-7-frankja@linux.ibm.com>
(cherry picked from commit c370d5300f9ac1f90f8158082d22262b904fe30e)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 22 ++++++++--------------
1 file changed, 8 insertions(+), 14 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index f6fe13e258..902a85ef8e 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1557,25 +1557,19 @@ bool dump_in_progress(void)
return (qatomic_read(&state->status) == DUMP_STATUS_ACTIVE);
}
-/* calculate total size of memory to be dumped (taking filter into
- * acoount.) */
+/*
+ * calculate total size of memory to be dumped (taking filter into
+ * account.)
+ */
static int64_t dump_calculate_size(DumpState *s)
{
GuestPhysBlock *block;
- int64_t size = 0, total = 0, left = 0, right = 0;
+ int64_t total = 0;
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- if (dump_has_filter(s)) {
- /* calculate the overlapped region. */
- left = MAX(s->filter_area_begin, block->target_start);
- right = MIN(s->filter_area_begin + s->filter_area_length, block->target_end);
- size = right - left;
- size = size > 0 ? size : 0;
- } else {
- /* count the whole region in */
- size = (block->target_end - block->target_start);
- }
- total += size;
+ total += dump_filtered_memblock_size(block,
+ s->filter_area_begin,
+ s->filter_area_length);
}
return total;
--
2.37.3

187
SOURCES/kvm-dump-Rework-filter-area-variables.patch
Unescape View File

@ -0,0 +1,187 @@
From 411f5354b809f6b783946e58d7655135814fb809 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:58 +0000
Subject: [PATCH 26/42] dump: Rework filter area variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [26/41] f10a5523dfd2724f7a8637fca3ed68ba6df659a5
While the DumpState begin and length variables directly mirror the API
variable names they are not very descriptive. So let's add a
"filter_area_" prefix and make has_filter a function checking length > 0.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-6-frankja@linux.ibm.com>
(cherry picked from commit dddf725f70bfe7f5adb41fa31dbd06e767271bda)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 53 +++++++++++++++++++++++++------------------
include/sysemu/dump.h | 13 ++++++++---
2 files changed, 41 insertions(+), 25 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index e6aa037f59..f6fe13e258 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -60,6 +60,11 @@ static inline bool dump_is_64bit(DumpState *s)
return s->dump_info.d_class == ELFCLASS64;
}
+static inline bool dump_has_filter(DumpState *s)
+{
+ return s->filter_area_length > 0;
+}
+
uint16_t cpu_to_dump16(DumpState *s, uint16_t val)
{
if (s->dump_info.d_endian == ELFDATA2LSB) {
@@ -444,29 +449,30 @@ static void get_offset_range(hwaddr phys_addr,
*p_offset = -1;
*p_filesz = 0;
- if (s->has_filter) {
- if (phys_addr < s->begin || phys_addr >= s->begin + s->length) {
+ if (dump_has_filter(s)) {
+ if (phys_addr < s->filter_area_begin ||
+ phys_addr >= s->filter_area_begin + s->filter_area_length) {
return;
}
}
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- if (s->has_filter) {
- if (block->target_start >= s->begin + s->length ||
- block->target_end <= s->begin) {
+ if (dump_has_filter(s)) {
+ if (block->target_start >= s->filter_area_begin + s->filter_area_length ||
+ block->target_end <= s->filter_area_begin) {
/* This block is out of the range */
continue;
}
- if (s->begin <= block->target_start) {
+ if (s->filter_area_begin <= block->target_start) {
start = block->target_start;
} else {
- start = s->begin;
+ start = s->filter_area_begin;
}
size_in_block = block->target_end - start;
- if (s->begin + s->length < block->target_end) {
- size_in_block -= block->target_end - (s->begin + s->length);
+ if (s->filter_area_begin + s->filter_area_length < block->target_end) {
+ size_in_block -= block->target_end - (s->filter_area_begin + s->filter_area_length);
}
} else {
start = block->target_start;
@@ -639,12 +645,12 @@ static void dump_iterate(DumpState *s, Error **errp)
int64_t memblock_size, memblock_start;
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- memblock_start = dump_filtered_memblock_start(block, s->begin, s->length);
+ memblock_start = dump_filtered_memblock_start(block, s->filter_area_begin, s->filter_area_length);
if (memblock_start == -1) {
continue;
}
- memblock_size = dump_filtered_memblock_size(block, s->begin, s->length);
+ memblock_size = dump_filtered_memblock_size(block, s->filter_area_begin, s->filter_area_length);
/* Write the memory to file */
write_memory(s, block, memblock_start, memblock_size, errp);
@@ -1513,14 +1519,14 @@ static int validate_start_block(DumpState *s)
{
GuestPhysBlock *block;
- if (!s->has_filter) {
+ if (!dump_has_filter(s)) {
return 0;
}
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
/* This block is out of the range */
- if (block->target_start >= s->begin + s->length ||
- block->target_end <= s->begin) {
+ if (block->target_start >= s->filter_area_begin + s->filter_area_length ||
+ block->target_end <= s->filter_area_begin) {
continue;
}
return 0;
@@ -1559,10 +1565,10 @@ static int64_t dump_calculate_size(DumpState *s)
int64_t size = 0, total = 0, left = 0, right = 0;
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- if (s->has_filter) {
+ if (dump_has_filter(s)) {
/* calculate the overlapped region. */
- left = MAX(s->begin, block->target_start);
- right = MIN(s->begin + s->length, block->target_end);
+ left = MAX(s->filter_area_begin, block->target_start);
+ right = MIN(s->filter_area_begin + s->filter_area_length, block->target_end);
size = right - left;
size = size > 0 ? size : 0;
} else {
@@ -1652,9 +1658,12 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
s->fd = fd;
- s->has_filter = has_filter;
- s->begin = begin;
- s->length = length;
+ if (has_filter && !length) {
+ error_setg(errp, QERR_INVALID_PARAMETER, "length");
+ goto cleanup;
+ }
+ s->filter_area_begin = begin;
+ s->filter_area_length = length;
memory_mapping_list_init(&s->list);
@@ -1787,8 +1796,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
return;
}
- if (s->has_filter) {
- memory_mapping_filter(&s->list, s->begin, s->length);
+ if (dump_has_filter(s)) {
+ memory_mapping_filter(&s->list, s->filter_area_begin, s->filter_area_length);
}
/*
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 7fce1d4af6..b62513d87d 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -166,9 +166,16 @@ typedef struct DumpState {
hwaddr memory_offset;
int fd;
- bool has_filter;
- int64_t begin;
- int64_t length;
+ /*
+ * Dump filter area variables
+ *
+ * A filtered dump only contains the guest memory designated by
+ * the start address and length variables defined below.
+ *
+ * If length is 0, no filtering is applied.
+ */
+ int64_t filter_area_begin; /* Start address of partial guest memory area */
+ int64_t filter_area_length; /* Length of partial guest memory area */
uint8_t *note_buf; /* buffer for notes */
size_t note_buf_offset; /* the writing place in note_buf */
--
2.37.3

102
SOURCES/kvm-dump-Rework-get_start_block.patch
Unescape View File

@ -0,0 +1,102 @@
From b56c362132baef40cc25d910c1e0d217d83cfe44 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:57 +0000
Subject: [PATCH 25/42] dump: Rework get_start_block
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [25/41] c93842a1aaeadcc11e91c194452fcd05d163b3ca
get_start_block() returns the start address of the first memory block
or -1.
With the GuestPhysBlock iterator conversion we don't need to set the
start address and can therefore remove that code and the "start"
DumpState struct member. The only functionality left is the validation
of the start block so it only makes sense to re-name the function to
validate_start_block()
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Message-Id: <20220811121111.9878-5-frankja@linux.ibm.com>
(cherry picked from commit 0c2994ac9009577b967529ce18e269da5b280351)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 20 ++++++--------------
include/sysemu/dump.h | 2 --
2 files changed, 6 insertions(+), 16 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index d981e843dd..e6aa037f59 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1509,30 +1509,22 @@ static void create_kdump_vmcore(DumpState *s, Error **errp)
}
}
-static ram_addr_t get_start_block(DumpState *s)
+static int validate_start_block(DumpState *s)
{
GuestPhysBlock *block;
if (!s->has_filter) {
- s->next_block = QTAILQ_FIRST(&s->guest_phys_blocks.head);
return 0;
}
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
+ /* This block is out of the range */
if (block->target_start >= s->begin + s->length ||
block->target_end <= s->begin) {
- /* This block is out of the range */
continue;
}
-
- s->next_block = block;
- if (s->begin > block->target_start) {
- s->start = s->begin - block->target_start;
- } else {
- s->start = 0;
- }
- return s->start;
- }
+ return 0;
+ }
return -1;
}
@@ -1679,8 +1671,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
goto cleanup;
}
- s->start = get_start_block(s);
- if (s->start == -1) {
+ /* Is the filter filtering everything? */
+ if (validate_start_block(s) == -1) {
error_setg(errp, QERR_INVALID_PARAMETER, "begin");
goto cleanup;
}
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index ffc2ea1072..7fce1d4af6 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -166,8 +166,6 @@ typedef struct DumpState {
hwaddr memory_offset;
int fd;
- GuestPhysBlock *next_block;
- ram_addr_t start;
bool has_filter;
int64_t begin;
int64_t length;
--
2.37.3

173
SOURCES/kvm-dump-Split-elf-header-functions-into-prepare-and-wri.patch
Unescape View File

@ -0,0 +1,173 @@
From d1e147a3133d4d31d4b0c02c05916366fadd9c30 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:11:00 +0000
Subject: [PATCH 28/42] dump: Split elf header functions into prepare and write
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [28/41] f70a13ad443835e7f46b7c5e176e372d370ac797
Let's split the write from the modification of the elf header so we
can consolidate the write of the data in one function.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-8-frankja@linux.ibm.com>
(cherry picked from commit 670e76998a61ca171200fcded3865b294a2d1243)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 100 ++++++++++++++++++++++++++++------------------------
1 file changed, 53 insertions(+), 47 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 902a85ef8e..8d5226f861 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -132,7 +132,7 @@ static int fd_write_vmcore(const void *buf, size_t size, void *opaque)
return 0;
}
-static void write_elf64_header(DumpState *s, Error **errp)
+static void prepare_elf64_header(DumpState *s, Elf64_Ehdr *elf_header)
{
/*
* phnum in the elf header is 16 bit, if we have more segments we
@@ -140,34 +140,27 @@ static void write_elf64_header(DumpState *s, Error **errp)
* special section.
*/
uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
- Elf64_Ehdr elf_header;
- int ret;
- memset(&elf_header, 0, sizeof(Elf64_Ehdr));
- memcpy(&elf_header, ELFMAG, SELFMAG);
- elf_header.e_ident[EI_CLASS] = ELFCLASS64;
- elf_header.e_ident[EI_DATA] = s->dump_info.d_endian;
- elf_header.e_ident[EI_VERSION] = EV_CURRENT;
- elf_header.e_type = cpu_to_dump16(s, ET_CORE);
- elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
- elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
- elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump64(s, s->phdr_offset);
- elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, phnum);
+ memset(elf_header, 0, sizeof(Elf64_Ehdr));
+ memcpy(elf_header, ELFMAG, SELFMAG);
+ elf_header->e_ident[EI_CLASS] = ELFCLASS64;
+ elf_header->e_ident[EI_DATA] = s->dump_info.d_endian;
+ elf_header->e_ident[EI_VERSION] = EV_CURRENT;
+ elf_header->e_type = cpu_to_dump16(s, ET_CORE);
+ elf_header->e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
+ elf_header->e_version = cpu_to_dump32(s, EV_CURRENT);
+ elf_header->e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
+ elf_header->e_phoff = cpu_to_dump64(s, s->phdr_offset);
+ elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
+ elf_header->e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- elf_header.e_shoff = cpu_to_dump64(s, s->shdr_offset);
- elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
- }
-
- ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
- if (ret < 0) {
- error_setg_errno(errp, -ret, "dump: failed to write elf header");
+ elf_header->e_shoff = cpu_to_dump64(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
}
}
-static void write_elf32_header(DumpState *s, Error **errp)
+static void prepare_elf32_header(DumpState *s, Elf32_Ehdr *elf_header)
{
/*
* phnum in the elf header is 16 bit, if we have more segments we
@@ -175,28 +168,45 @@ static void write_elf32_header(DumpState *s, Error **errp)
* special section.
*/
uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
- Elf32_Ehdr elf_header;
- int ret;
- memset(&elf_header, 0, sizeof(Elf32_Ehdr));
- memcpy(&elf_header, ELFMAG, SELFMAG);
- elf_header.e_ident[EI_CLASS] = ELFCLASS32;
- elf_header.e_ident[EI_DATA] = s->dump_info.d_endian;
- elf_header.e_ident[EI_VERSION] = EV_CURRENT;
- elf_header.e_type = cpu_to_dump16(s, ET_CORE);
- elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
- elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
- elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump32(s, s->phdr_offset);
- elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, phnum);
+ memset(elf_header, 0, sizeof(Elf32_Ehdr));
+ memcpy(elf_header, ELFMAG, SELFMAG);
+ elf_header->e_ident[EI_CLASS] = ELFCLASS32;
+ elf_header->e_ident[EI_DATA] = s->dump_info.d_endian;
+ elf_header->e_ident[EI_VERSION] = EV_CURRENT;
+ elf_header->e_type = cpu_to_dump16(s, ET_CORE);
+ elf_header->e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
+ elf_header->e_version = cpu_to_dump32(s, EV_CURRENT);
+ elf_header->e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
+ elf_header->e_phoff = cpu_to_dump32(s, s->phdr_offset);
+ elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
+ elf_header->e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- elf_header.e_shoff = cpu_to_dump32(s, s->shdr_offset);
- elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
+ elf_header->e_shoff = cpu_to_dump32(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
}
+}
- ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
+static void write_elf_header(DumpState *s, Error **errp)
+{
+ Elf32_Ehdr elf32_header;
+ Elf64_Ehdr elf64_header;
+ size_t header_size;
+ void *header_ptr;
+ int ret;
+
+ if (dump_is_64bit(s)) {
+ prepare_elf64_header(s, &elf64_header);
+ header_size = sizeof(elf64_header);
+ header_ptr = &elf64_header;
+ } else {
+ prepare_elf32_header(s, &elf32_header);
+ header_size = sizeof(elf32_header);
+ header_ptr = &elf32_header;
+ }
+
+ ret = fd_write_vmcore(header_ptr, header_size, s);
if (ret < 0) {
error_setg_errno(errp, -ret, "dump: failed to write elf header");
}
@@ -565,11 +575,7 @@ static void dump_begin(DumpState *s, Error **errp)
*/
/* write elf header to vmcore */
- if (dump_is_64bit(s)) {
- write_elf64_header(s, errp);
- } else {
- write_elf32_header(s, errp);
- }
+ write_elf_header(s, errp);
if (*errp) {
return;
}
--
2.37.3

420
SOURCES/kvm-dump-Use-ERRP_GUARD.patch
Unescape View File

@ -0,0 +1,420 @@
From 4ca61efe246d62d420eb332655c0c8ead4cc762b Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:55 +0000
Subject: [PATCH 13/42] dump: Use ERRP_GUARD()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [13/41] f735cd1dab0230000cfadd878765fdf4647b239c
Let's move to the new way of handling errors before changing the dump
code. This patch has mostly been generated by the coccinelle script
scripts/coccinelle/errp-guard.cocci.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-2-frankja@linux.ibm.com>
(cherry picked from commit 86a518bba4f4d7c9016fc5b104fe1e58b00ad756)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 144 ++++++++++++++++++++++------------------------------
1 file changed, 61 insertions(+), 83 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 662d0a62cd..9876123f2e 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -390,23 +390,21 @@ static void write_data(DumpState *s, void *buf, int length, Error **errp)
static void write_memory(DumpState *s, GuestPhysBlock *block, ram_addr_t start,
int64_t size, Error **errp)
{
+ ERRP_GUARD();
int64_t i;
- Error *local_err = NULL;
for (i = 0; i < size / s->dump_info.page_size; i++) {
write_data(s, block->host_addr + start + i * s->dump_info.page_size,
- s->dump_info.page_size, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ s->dump_info.page_size, errp);
+ if (*errp) {
return;
}
}
if ((size % s->dump_info.page_size) != 0) {
write_data(s, block->host_addr + start + i * s->dump_info.page_size,
- size % s->dump_info.page_size, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ size % s->dump_info.page_size, errp);
+ if (*errp) {
return;
}
}
@@ -476,11 +474,11 @@ static void get_offset_range(hwaddr phys_addr,
static void write_elf_loads(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
hwaddr offset, filesz;
MemoryMapping *memory_mapping;
uint32_t phdr_index = 1;
uint32_t max_index;
- Error *local_err = NULL;
if (s->have_section) {
max_index = s->sh_info;
@@ -494,14 +492,13 @@ static void write_elf_loads(DumpState *s, Error **errp)
s, &offset, &filesz);
if (s->dump_info.d_class == ELFCLASS64) {
write_elf64_load(s, memory_mapping, phdr_index++, offset,
- filesz, &local_err);
+ filesz, errp);
} else {
write_elf32_load(s, memory_mapping, phdr_index++, offset,
- filesz, &local_err);
+ filesz, errp);
}
- if (local_err) {
- error_propagate(errp, local_err);
+ if (*errp) {
return;
}
@@ -514,7 +511,7 @@ static void write_elf_loads(DumpState *s, Error **errp)
/* write elf header, PT_NOTE and elf note to vmcore. */
static void dump_begin(DumpState *s, Error **errp)
{
- Error *local_err = NULL;
+ ERRP_GUARD();
/*
* the vmcore's format is:
@@ -542,73 +539,64 @@ static void dump_begin(DumpState *s, Error **errp)
/* write elf header to vmcore */
if (s->dump_info.d_class == ELFCLASS64) {
- write_elf64_header(s, &local_err);
+ write_elf64_header(s, errp);
} else {
- write_elf32_header(s, &local_err);
+ write_elf32_header(s, errp);
}
- if (local_err) {
- error_propagate(errp, local_err);
+ if (*errp) {
return;
}
if (s->dump_info.d_class == ELFCLASS64) {
/* write PT_NOTE to vmcore */
- write_elf64_note(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf64_note(s, errp);
+ if (*errp) {
return;
}
/* write all PT_LOAD to vmcore */
- write_elf_loads(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_loads(s, errp);
+ if (*errp) {
return;
}
/* write section to vmcore */
if (s->have_section) {
- write_elf_section(s, 1, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_section(s, 1, errp);
+ if (*errp) {
return;
}
}
/* write notes to vmcore */
- write_elf64_notes(fd_write_vmcore, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf64_notes(fd_write_vmcore, s, errp);
+ if (*errp) {
return;
}
} else {
/* write PT_NOTE to vmcore */
- write_elf32_note(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf32_note(s, errp);
+ if (*errp) {
return;
}
/* write all PT_LOAD to vmcore */
- write_elf_loads(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_loads(s, errp);
+ if (*errp) {
return;
}
/* write section to vmcore */
if (s->have_section) {
- write_elf_section(s, 0, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_section(s, 0, errp);
+ if (*errp) {
return;
}
}
/* write notes to vmcore */
- write_elf32_notes(fd_write_vmcore, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf32_notes(fd_write_vmcore, s, errp);
+ if (*errp) {
return;
}
}
@@ -644,9 +632,9 @@ static int get_next_block(DumpState *s, GuestPhysBlock *block)
/* write all memory to vmcore */
static void dump_iterate(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
GuestPhysBlock *block;
int64_t size;
- Error *local_err = NULL;
do {
block = s->next_block;
@@ -658,9 +646,8 @@ static void dump_iterate(DumpState *s, Error **errp)
size -= block->target_end - (s->begin + s->length);
}
}
- write_memory(s, block, s->start, size, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_memory(s, block, s->start, size, errp);
+ if (*errp) {
return;
}
@@ -669,11 +656,10 @@ static void dump_iterate(DumpState *s, Error **errp)
static void create_vmcore(DumpState *s, Error **errp)
{
- Error *local_err = NULL;
+ ERRP_GUARD();
- dump_begin(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ dump_begin(s, errp);
+ if (*errp) {
return;
}
@@ -810,6 +796,7 @@ static bool note_name_equal(DumpState *s,
/* write common header, sub header and elf note to vmcore */
static void create_header32(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
DiskDumpHeader32 *dh = NULL;
KdumpSubHeader32 *kh = NULL;
size_t size;
@@ -818,7 +805,6 @@ static void create_header32(DumpState *s, Error **errp)
uint32_t bitmap_blocks;
uint32_t status = 0;
uint64_t offset_note;
- Error *local_err = NULL;
/* write common header, the version of kdump-compressed format is 6th */
size = sizeof(DiskDumpHeader32);
@@ -894,9 +880,8 @@ static void create_header32(DumpState *s, Error **errp)
s->note_buf_offset = 0;
/* use s->note_buf to store notes temporarily */
- write_elf32_notes(buf_write_note, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf32_notes(buf_write_note, s, errp);
+ if (*errp) {
goto out;
}
if (write_buffer(s->fd, offset_note, s->note_buf,
@@ -922,6 +907,7 @@ out:
/* write common header, sub header and elf note to vmcore */
static void create_header64(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
DiskDumpHeader64 *dh = NULL;
KdumpSubHeader64 *kh = NULL;
size_t size;
@@ -930,7 +916,6 @@ static void create_header64(DumpState *s, Error **errp)
uint32_t bitmap_blocks;
uint32_t status = 0;
uint64_t offset_note;
- Error *local_err = NULL;
/* write common header, the version of kdump-compressed format is 6th */
size = sizeof(DiskDumpHeader64);
@@ -1006,9 +991,8 @@ static void create_header64(DumpState *s, Error **errp)
s->note_buf_offset = 0;
/* use s->note_buf to store notes temporarily */
- write_elf64_notes(buf_write_note, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf64_notes(buf_write_note, s, errp);
+ if (*errp) {
goto out;
}
@@ -1472,8 +1456,8 @@ out:
static void create_kdump_vmcore(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
int ret;
- Error *local_err = NULL;
/*
* the kdump-compressed format is:
@@ -1503,21 +1487,18 @@ static void create_kdump_vmcore(DumpState *s, Error **errp)
return;
}
- write_dump_header(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_dump_header(s, errp);
+ if (*errp) {
return;
}
- write_dump_bitmap(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_dump_bitmap(s, errp);
+ if (*errp) {
return;
}
- write_dump_pages(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_dump_pages(s, errp);
+ if (*errp) {
return;
}
@@ -1647,10 +1628,10 @@ static void dump_init(DumpState *s, int fd, bool has_format,
DumpGuestMemoryFormat format, bool paging, bool has_filter,
int64_t begin, int64_t length, Error **errp)
{
+ ERRP_GUARD();
VMCoreInfoState *vmci = vmcoreinfo_find();
CPUState *cpu;
int nr_cpus;
- Error *err = NULL;
int ret;
s->has_format = has_format;
@@ -1769,9 +1750,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
/* get memory mapping */
if (paging) {
- qemu_get_guest_memory_mapping(&s->list, &s->guest_phys_blocks, &err);
- if (err != NULL) {
- error_propagate(errp, err);
+ qemu_get_guest_memory_mapping(&s->list, &s->guest_phys_blocks, errp);
+ if (*errp) {
goto cleanup;
}
} else {
@@ -1870,33 +1850,32 @@ cleanup:
/* this operation might be time consuming. */
static void dump_process(DumpState *s, Error **errp)
{
- Error *local_err = NULL;
+ ERRP_GUARD();
DumpQueryResult *result = NULL;
if (s->has_format && s->format == DUMP_GUEST_MEMORY_FORMAT_WIN_DMP) {
#ifdef TARGET_X86_64
- create_win_dump(s, &local_err);
+ create_win_dump(s, errp);
#endif
} else if (s->has_format && s->format != DUMP_GUEST_MEMORY_FORMAT_ELF) {
- create_kdump_vmcore(s, &local_err);
+ create_kdump_vmcore(s, errp);
} else {
- create_vmcore(s, &local_err);
+ create_vmcore(s, errp);
}
/* make sure status is written after written_size updates */
smp_wmb();
qatomic_set(&s->status,
- (local_err ? DUMP_STATUS_FAILED : DUMP_STATUS_COMPLETED));
+ (*errp ? DUMP_STATUS_FAILED : DUMP_STATUS_COMPLETED));
/* send DUMP_COMPLETED message (unconditionally) */
result = qmp_query_dump(NULL);
/* should never fail */
assert(result);
- qapi_event_send_dump_completed(result, !!local_err, (local_err ?
- error_get_pretty(local_err) : NULL));
+ qapi_event_send_dump_completed(result, !!*errp, (*errp ?
+ error_get_pretty(*errp) : NULL));
qapi_free_DumpQueryResult(result);
- error_propagate(errp, local_err);
dump_cleanup(s);
}
@@ -1925,10 +1904,10 @@ void qmp_dump_guest_memory(bool paging, const char *file,
int64_t length, bool has_format,
DumpGuestMemoryFormat format, Error **errp)
{
+ ERRP_GUARD();
const char *p;
int fd = -1;
DumpState *s;
- Error *local_err = NULL;
bool detach_p = false;
if (runstate_check(RUN_STATE_INMIGRATE)) {
@@ -2028,9 +2007,8 @@ void qmp_dump_guest_memory(bool paging, const char *file,
dump_state_prepare(s);
dump_init(s, fd, has_format, format, paging, has_begin,
- begin, length, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ begin, length, errp);
+ if (*errp) {
qatomic_set(&s->status, DUMP_STATUS_FAILED);
return;
}
--
2.37.3

150
SOURCES/kvm-dump-Use-a-buffer-for-ELF-section-data-and-headers.patch
Unescape View File

@ -0,0 +1,150 @@
From a918c7305ec7c68e8bc37b449f71e75d84124cd0 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:13 +0000
Subject: [PATCH 32/42] dump: Use a buffer for ELF section data and headers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [32/41] e1a03e202e67764581e486f37e13e479200e5846
Currently we're writing the NULL section header if we overflow the
physical header number in the ELF header. But in the future we'll add
custom section headers AND section data.
To facilitate this we need to rearange section handling a bit. As with
the other ELF headers we split the code into a prepare and a write
step.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-2-frankja@linux.ibm.com>
(cherry picked from commit e41ed29bcee5cb16715317bcf290f6b5c196eb0a)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 75 +++++++++++++++++++++++++++++--------------
include/sysemu/dump.h | 2 ++
2 files changed, 53 insertions(+), 24 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 88177fa886..4142b4cc0c 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -381,31 +381,60 @@ static void write_elf_phdr_note(DumpState *s, Error **errp)
}
}
-static void write_elf_section(DumpState *s, int type, Error **errp)
+static void prepare_elf_section_hdr_zero(DumpState *s)
{
- Elf32_Shdr shdr32;
- Elf64_Shdr shdr64;
- int shdr_size;
- void *shdr;
- int ret;
+ if (dump_is_64bit(s)) {
+ Elf64_Shdr *shdr64 = s->elf_section_hdrs;
- if (type == 0) {
- shdr_size = sizeof(Elf32_Shdr);
- memset(&shdr32, 0, shdr_size);
- shdr32.sh_info = cpu_to_dump32(s, s->phdr_num);
- shdr = &shdr32;
+ shdr64->sh_info = cpu_to_dump32(s, s->phdr_num);
} else {
- shdr_size = sizeof(Elf64_Shdr);
- memset(&shdr64, 0, shdr_size);
- shdr64.sh_info = cpu_to_dump32(s, s->phdr_num);
- shdr = &shdr64;
+ Elf32_Shdr *shdr32 = s->elf_section_hdrs;
+
+ shdr32->sh_info = cpu_to_dump32(s, s->phdr_num);
+ }
+}
+
+static void prepare_elf_section_hdrs(DumpState *s)
+{
+ size_t len, sizeof_shdr;
+
+ /*
+ * Section ordering:
+ * - HDR zero
+ */
+ sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
+ len = sizeof_shdr * s->shdr_num;
+ s->elf_section_hdrs = g_malloc0(len);
+
+ /*
+ * The first section header is ALWAYS a special initial section
+ * header.
+ *
+ * The header should be 0 with one exception being that if
+ * phdr_num is PN_XNUM then the sh_info field contains the real
+ * number of segment entries.
+ *
+ * As we zero allocate the buffer we will only need to modify
+ * sh_info for the PN_XNUM case.
+ */
+ if (s->phdr_num >= PN_XNUM) {
+ prepare_elf_section_hdr_zero(s);
}
+}
- ret = fd_write_vmcore(shdr, shdr_size, s);
+static void write_elf_section_headers(DumpState *s, Error **errp)
+{
+ size_t sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
+ int ret;
+
+ prepare_elf_section_hdrs(s);
+
+ ret = fd_write_vmcore(s->elf_section_hdrs, s->shdr_num * sizeof_shdr, s);
if (ret < 0) {
- error_setg_errno(errp, -ret,
- "dump: failed to write section header table");
+ error_setg_errno(errp, -ret, "dump: failed to write section headers");
}
+
+ g_free(s->elf_section_hdrs);
}
static void write_data(DumpState *s, void *buf, int length, Error **errp)
@@ -592,12 +621,10 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- /* write section to vmcore */
- if (s->shdr_num) {
- write_elf_section(s, 1, errp);
- if (*errp) {
- return;
- }
+ /* write section headers to vmcore */
+ write_elf_section_headers(s, errp);
+ if (*errp) {
+ return;
}
/* write notes to vmcore */
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index b62513d87d..9995f65dc8 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -177,6 +177,8 @@ typedef struct DumpState {
int64_t filter_area_begin; /* Start address of partial guest memory area */
int64_t filter_area_length; /* Length of partial guest memory area */
+ void *elf_section_hdrs; /* Pointer to section header buffer */
+
uint8_t *note_buf; /* buffer for notes */
size_t note_buf_offset; /* the writing place in note_buf */
uint32_t nr_cpus; /* number of guest's cpu */
--
2.37.3

104
SOURCES/kvm-dump-Write-ELF-section-headers-right-after-ELF-heade.patch
Unescape View File

@ -0,0 +1,104 @@
From 987ede93fa4e3d058acddc19874e467faa116ede Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:14 +0000
Subject: [PATCH 33/42] dump: Write ELF section headers right after ELF header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [33/41] e956040753533ac376e9763145192de1e216027d
Let's start bundling the writes of the headers and of the data so we
have a clear ordering between them. Since the ELF header uses offsets
to the headers we can freely order them.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-3-frankja@linux.ibm.com>
(cherry picked from commit cb415fd61e48d52f81dcf38956e3f913651cff1c)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 31 ++++++++++++++-----------------
1 file changed, 14 insertions(+), 17 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 4142b4cc0c..d17537d4e9 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -584,6 +584,8 @@ static void dump_begin(DumpState *s, Error **errp)
* --------------
* | elf header |
* --------------
+ * | sctn_hdr |
+ * --------------
* | PT_NOTE |
* --------------
* | PT_LOAD |
@@ -592,8 +594,6 @@ static void dump_begin(DumpState *s, Error **errp)
* --------------
* | PT_LOAD |
* --------------
- * | sec_hdr |
- * --------------
* | elf note |
* --------------
* | memory |
@@ -609,20 +609,20 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- /* write PT_NOTE to vmcore */
- write_elf_phdr_note(s, errp);
+ /* write section headers to vmcore */
+ write_elf_section_headers(s, errp);
if (*errp) {
return;
}
- /* write all PT_LOADs to vmcore */
- write_elf_phdr_loads(s, errp);
+ /* write PT_NOTE to vmcore */
+ write_elf_phdr_note(s, errp);
if (*errp) {
return;
}
- /* write section headers to vmcore */
- write_elf_section_headers(s, errp);
+ /* write all PT_LOADs to vmcore */
+ write_elf_phdr_loads(s, errp);
if (*errp) {
return;
}
@@ -1877,16 +1877,13 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (dump_is_64bit(s)) {
- s->phdr_offset = sizeof(Elf64_Ehdr);
- s->shdr_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
- s->note_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
- s->memory_offset = s->note_offset + s->note_size;
+ s->shdr_offset = sizeof(Elf64_Ehdr);
+ s->phdr_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
+ s->note_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
} else {
-
- s->phdr_offset = sizeof(Elf32_Ehdr);
- s->shdr_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
- s->note_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
- s->memory_offset = s->note_offset + s->note_size;
+ s->shdr_offset = sizeof(Elf32_Ehdr);
+ s->phdr_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
+ s->note_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
}
return;
--
2.37.3

Some files were not shown because too many files have changed in this diff Show More

Write Preview
Loading…
Cancel
Save