[edk2-devel] [PATCH v3 0/2] OvmfPkg/PlatformInitLib: catch QEMU's CPU hotplug reg block regression

[Laszlo Ersek]

Repo:       https://pagure.io/lersek/edk2.git
Branch:     cpuhp-reg-catch-4250-v3
Test build: https://github.com/tianocore/edk2/pull/3930
Bugzilla:   https://bugzilla.tianocore.org/show_bug.cgi?id=4250

v2 was posted at:
- http://mid.mail-archive.com/20230112082845.128463-1-lersek@redhat.com
- https://edk2.groups.io/g/devel/message/98336
- https://listman.redhat.com/archives/edk2-devel-archive/2023-January/057634.html

Please see the Notes sections of the patches, regarding the updates.

The "PlatformCI_OvmfPkg_Windows_VS2019_PR" checks in test build linked
above time out again (as expected); now with the following debug log:

> PlatformCpuCountBugCheck: Present=0 Possible=1
> PlatformCpuCountBugCheck: Broken CPU hotplug register block found. Update QEMU to version 8+, or
> PlatformCpuCountBugCheck: to a stable release with commit dab30fbef389 backported. Refer to
> PlatformCpuCountBugCheck: <https://bugzilla.tianocore.org/show_bug.cgi?id=4250>.
> PlatformCpuCountBugCheck: Consequences of the QEMU bug may include, but are not limited to:
> PlatformCpuCountBugCheck: - all firmware logic, dependent on the CPU hotplug register block,
> PlatformCpuCountBugCheck:   being confused, for example, multiprocessing-related logic;
> PlatformCpuCountBugCheck: - guest OS data loss, including filesystem corruption, due to crash or
> PlatformCpuCountBugCheck:   hang during ACPI S3 resume;
> PlatformCpuCountBugCheck: - SMM privilege escalation, by a malicious guest OS or 3rd partty UEFI
> PlatformCpuCountBugCheck:   agent, against the platform firmware.
> PlatformCpuCountBugCheck: These symptoms need not necessarily be limited to the QEMU user
> PlatformCpuCountBugCheck: attempting to hot(un)plug a CPU.
> PlatformCpuCountBugCheck: The firmware will now stop (hang) deliberately, in order to prevent the
> PlatformCpuCountBugCheck: above symptoms.
> PlatformCpuCountBugCheck: You can forcibly override the hang, *at your own risk*, with the
> PlatformCpuCountBugCheck: following *experimental* QEMU command line option:
> PlatformCpuCountBugCheck:   -fw_cfg name=opt/org.tianocore/X-Cpuhp-Bugcheck-Override,string=yes
> PlatformCpuCountBugCheck: Please only report such bugs that you can reproduce *without* the
> PlatformCpuCountBugCheck: override.
> Select Item: 0x19
> ASSERT d:\a\1\s\OvmfPkg\Library\PlatformInitLib\Platform.c(520): ((BOOLEAN)(0==1))

The "PlatformCI_OvmfPkg_Ubuntu_GCC5_PR" checks succed, probably due to
edk2 commits ef0916009843 ("CI: Use Fedora 35 container (Linux only)",
2023-01-17) and 7fab007f33e9 ("OvmfPkg: CI: Use Fedora 35 container
(Linux only)", 2023-01-17), and due to
<https://github.com/tianocore/containers/commit/47addc9a4f20> applying
the upstream QEMU patch.

Laszlo

Cc: Ard Biesheuvel <ardb+tianocore at kernel.org>
Cc: Brijesh Singh <brijesh.singh at amd.com>
Cc: Erdem Aktas <erdemaktas at google.com>
Cc: Gerd Hoffmann <kraxel at redhat.com>
Cc: James Bottomley <jejb at linux.ibm.com>
Cc: Jiewen Yao <jiewen.yao at intel.com>
Cc: Jordan Justen <jordan.l.justen at intel.com>
Cc: Michael Brown <mcb30 at ipxe.org>
Cc: Min Xu <min.m.xu at intel.com>
Cc: Oliver Steffen <osteffen at redhat.com>
Cc: Sebastien Boeuf <sebastien.boeuf at intel.com>
Cc: Tom Lendacky <thomas.lendacky at amd.com>

Laszlo Ersek (2):
  OvmfPkg/PlatformInitLib: factor out PlatformCpuCountBugCheck()
  OvmfPkg/PlatformInitLib: catch QEMU's CPU hotplug reg block regression

 OvmfPkg/Library/PlatformInitLib/Platform.c | 168 +++++++++++++++++---
 1 file changed, 145 insertions(+), 23 deletions(-)



-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#98886): https://edk2.groups.io/g/devel/message/98886
Mute This Topic: https://groups.io/mt/96374972/1813853
Group Owner: devel+owner at edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [edk2-devel-archive at redhat.com]
-=-=-=-=-=-=-=-=-=-=-=-