linux/arch/x86
Vitaly Kuznetsov d176620277 x86/kvm/lapic: always disable MMIO interface in x2APIC mode
When VMX is used with flexpriority disabled (because of no support or
if disabled with module parameter) MMIO interface to lAPIC is still
available in x2APIC mode while it shouldn't be (kvm-unit-tests):

PASS: apic_disable: Local apic enabled in x2APIC mode
PASS: apic_disable: CPUID.1H:EDX.APIC[bit 9] is set
FAIL: apic_disable: *0xfee00030: 50014

The issue appears because we basically do nothing while switching to
x2APIC mode when APIC access page is not used. apic_mmio_{read,write}
only check if lAPIC is disabled before proceeding to actual write.

When APIC access is virtualized we correctly manipulate with VMX controls
in vmx_set_virtual_apic_mode() and we don't get vmexits from memory writes
in x2APIC mode so there's no issue.

Disabling MMIO interface seems to be easy. The question is: what do we
do with these reads and writes? If we add apic_x2apic_mode() check to
apic_mmio_in_range() and return -EOPNOTSUPP these reads and writes will
go to userspace. When lAPIC is in kernel, Qemu uses this interface to
inject MSIs only (see kvm_apic_mem_write() in hw/i386/kvm/apic.c). This
somehow works with disabled lAPIC but when we're in xAPIC mode we will
get a real injected MSI from every write to lAPIC. Not good.

The simplest solution seems to be to just ignore writes to the region
and return ~0 for all reads when we're in x2APIC mode. This is what this
patch does. However, this approach is inconsistent with what currently
happens when flexpriority is enabled: we allocate APIC access page and
create KVM memory region so in x2APIC modes all reads and writes go to
this pre-allocated page which is, btw, the same for all vCPUs.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2018-09-20 00:26:43 +02:00
..
boot Kbuild updates for v4.19 (2nd) 2018-08-25 13:40:38 -07:00
configs
crypto Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6 2018-08-29 13:38:39 -07:00
entry x86/vdso: Fix vDSO build if a retpoline is emitted 2018-08-20 18:04:41 +02:00
events perf/x86/intel: Add support/quirk for the MISPREDICT bit on Knights Landing CPUs 2018-09-10 10:03:01 +02:00
hyperv x86/mm: Only use tlb_remove_table() for paravirt 2018-08-23 11:56:31 -07:00
ia32
include x86/kvm/lapic: always disable MMIO interface in x2APIC mode 2018-09-20 00:26:43 +02:00
kernel x86/APM: Fix build warning when PROC_FS is not enabled 2018-09-15 10:16:25 +02:00
kvm x86/kvm/lapic: always disable MMIO interface in x2APIC mode 2018-09-20 00:26:43 +02:00
lib x86/nmi: Fix NMI uaccess race against CR3 switching 2018-08-31 17:08:22 +02:00
math-emu
mm x86/mm: Use WRITE_ONCE() when setting PTEs 2018-09-08 12:30:36 +02:00
net bpf, x32: Fix regression caused by commit 24dea04767 2018-07-26 02:51:12 +02:00
oprofile
pci PCI: Make early dump functionality generic 2018-06-29 20:06:07 -05:00
platform x86/efi: Load fixmap GDT in efi_call_phys_epilog() before setting %cr3 2018-09-12 21:53:34 +02:00
power Power management updates for 4.19-rc1 2018-08-14 13:12:24 -07:00
purgatory kbuild: move bin2c back to scripts/ from scripts/basic/ 2018-07-18 01:18:05 +09:00
ras
realmode
tools x86/relocs: Add __end_rodata_aligned to S_REL 2018-08-09 20:42:07 +02:00
um Consolidation of Kconfig files by Christoph Hellwig. 2018-08-15 13:05:12 -07:00
video
xen xen: fixes for 4.19-rc2 2018-08-31 08:45:16 -07:00
.gitignore
Kbuild
Kconfig x86/Kconfig: Fix trivial typo 2018-08-27 10:29:14 +02:00
Kconfig.cpu
Kconfig.debug Kconfig: consolidate the "Kernel hacking" menu 2018-08-02 08:06:48 +09:00
Makefile x86: Allow generating user-space headers without a compiler 2018-08-31 17:08:22 +02:00
Makefile_32.cpu
Makefile.um kbuild: rename LDFLAGS to KBUILD_LDFLAGS 2018-08-24 08:22:08 +09:00