The current implementation has three bugs,
* segment limits are not enforced in protected mode if the L bit is set
in the target segment descriptor
* segment limits are not enforced in compatibility mode (ljmp to 32-bit
code segment in long mode)
* #GP(new_cs) is generated rather than #GP(0)
Now the segment limits are enforced if we're not in long mode OR the
target code segment doesn't have the L bit set.
Backports commit db7196db5d5d932f388643baae6835f8dcda6921 from qemu
Currently call gates are always treated as 32-bit gates. In IA-32e mode
(either compatibility or 64-bit submode), system segment descriptors are
always 64-bit. Treating them as 32-bit has the expected unfortunate
effect: only the lower 32 bits of the offset are loaded, the stack
pointer is truncated, a bad new stack pointer is loaded from the TSS (if
switching privilege levels), etc.
This change adds support for 64-bit call gate to the lcall and ljmp
instructions. Additionally, there should be a check for non-canonical
stack pointers, but I've omitted that since there doesn't seem to be
checks for non-canonical addresses in this code elsewhere.
I've left the raise_exception_err_ra lines unwapped at 80 columns to
match the style in the rest of the file.
Backports commit 0aca060526d3ff9632aaed66e8611814580c13de from qemu
Enabling TOPOEXT is always allowed, but it can't be enabled
blindly by "-cpu host" because it may make guests crash if the
rest of the cache topology information isn't provided or isn't
consistent.
This addresses the bug reported at:
https://bugzilla.redhat.com/show_bug.cgi?id=1613277
Backports commit 7210a02c58572b2686a3a8d610c6628f87864aed from qemu
New CPU models mostly inherit features from ancestor Skylake, while addin new
features: UMIP, New Instructions ( PCONIFIG (server only), WBNOINVD,
AVX512_VBMI2, GFNI, AVX512_VNNI, VPCLMULQDQ, VAES, AVX512_BITALG),
Intel PT and 5-level paging (Server only). As well as
IA32_PRED_CMD, SSBD support for speculative execution
side channel mitigations.
Note:
For 5-level paging, Guest physical address width can be configured, with
parameter "phys-bits". Unless explicitly specified, we still use its default
value, even for Icelake-Server cpu model.
At present, hold on expose IA32_ARCH_CAPABILITIES to guest, as 1) This MSR
actually presents more than 1 'feature', maintainers are considering expanding current
features presentation of only CPUIDs to MSR bits; 2) a reasonable default value
for MSR_IA32_ARCH_CAPABILITIES needs to settled first. These 2 are actully
beyond Icelake CPU model itself but fundamental. So split these work apart
and do it later.
https://lists.gnu.org/archive/html/qemu-devel/2018-07/msg00774.htmlhttps://lists.gnu.org/archive/html/qemu-devel/2018-07/msg00796.html
Backports commit 8a11c62da9146dd89aee98947e6bd831e65a970d from qemu
WBNOINVD: Write back and do not invalidate cache, enumerated by
CPUID.(EAX=80000008H, ECX=0):EBX[bit 9].
Backports commit 59a80a19ca31a6fff9fdbb6b4cf55a5a0767c3bc from qemu
Support of IA32_PRED_CMD MSR already be enumerated by same CPUID bit as
SPEC_CTRL.
At present, mark CPUID_7_0_EDX_ARCH_CAPABILITIES unmigratable, per Paolo's
comment.
Backports commit 3fc7c73139d2d38ae80c3b0bc963b1ac1555924c from qemu
IA32_PRED_CMD MSR gives software a way to issue commands that affect the state
of indirect branch predictors. Enumerated by CPUID.(EAX=7H,ECX=0):EDX[26].
IA32_ARCH_CAPABILITIES MSR enumerates architectural features of RDCL_NO and
IBRS_ALL. Enumerated by CPUID.(EAX=07H, ECX=0):EDX[29].
https://software.intel.com/sites/default/files/managed/c5/63/336996-Speculative-Execution-Side-Channel-Mitigations.pdf
Backports commit 8c80c99fcceabd0708a5a83f08577e778c9419f5 from qemu
MSR_SMI_COUNT started being migrated in QEMU 2.12. Do not migrate it
on older machine types, or the subsection causes a load failure for
guests that use SMM.
Backports part of commit 990e0be2603511560168e1ad61f68294d951c39e from
qemu
Rename DCACHE to DATA_CACHE and ICACHE to INSTRUCTION_CACHE.
This avoids conflict with Linux asm/cachectl.h macros and fixes
build failure on mips hosts.
Backports commit 5f00335aecafc9ad56592d943619d3252f8941f1 from qemu
This implements NPT suport for SVM by hooking into
x86_cpu_handle_mmu_fault where it reads the stage-1 page table. Whether
we need to perform this 2nd stage translation, and how, is decided
during vmrun and stored in hflags2, along with nested_cr3 and
nested_pg_mode.
As get_hphys performs a direct cpu_vmexit in case of NPT faults, we need
retaddr in that function. To avoid changing the signature of
cpu_handle_mmu_fault, this passes the value from tlb_fill to get_hphys
via the CPU state.
This was tested successfully via the Jailhouse hypervisor.
Backports commit fe441054bb3f0c75ff23335790342c0408e11c3a from qemu
It calls cpu_loop_exit in system emulation mode (and should never be
called in user emulation mode).
Backports commit 50b3de6e5cd464dcc20e3a48f5a09e0299a184ac from qemu
We need to terminate the translation block after STGI so that pending
interrupts can be injected.
This fixes pending NMI injection for Jailhouse which uses "stgi; clgi"
to open a brief injection window.
Backports commit df2518aa587a0157bbfbc635fe47295629d9914a from qemu
Check for SVM interception prior to injecting an NMI. Tested via the
Jailhouse hypervisor.
Backports commit 02f7fd25a446a220905c2e5cb0fc3655d7f63b29 from qemu
The implementation of these two instructions was swapped.
At the same time, unify the setup of eflags for the insn group.
Backports commit 13672386a93fef64cfd33bd72fbf3d80f2c00e94 from qemu
Enable TOPOEXT feature on EPYC CPU. This is required to support
hyperthreading on VM guests. Also extend xlevel to 0x8000001E.
Disable topoext on PC_COMPAT_2_12 and keep xlevel 0x8000000a.
Backports commit e00516475c270dcb6705753da96063f95699abf2 from qemu
This is part of topoext support. To keep the compatibility, it is better
we support all the combination of nr_cores and nr_threads currently
supported. By allowing more nr_cores and nr_threads, we might end up with
more nodes than we can actually support with the real hardware. We need to
fix up the node id to make this work. We can achieve this by shifting the
socket_id bits left to address more nodes.
Backports commit 631be32155dbafa1fe886f2488127956c9120ba6 from qemu
AMD future CPUs expose a mechanism to tell the guest that the
Speculative Store Bypass Disable is not needed and that the
CPU is all good.
This is exposed via the CPUID 8000_0008.EBX[26] bit.
See 124441_AMD64_SpeculativeStoreBypassDisable_Whitepaper_final.pdf
A copy of this document is available at
https://bugzilla.kernel.org/show_bug.cgi?id=199889
Backports commit 254790a909a2f153d689bfa7d8e8f0386cda870d from qemu
AMD future CPUs expose _two_ ways to utilize the Intel equivalant
of the Speculative Store Bypass Disable. The first is via
the virtualized VIRT_SPEC CTRL MSR (0xC001_011f) and the second
is via the SPEC_CTRL MSR (0x48). The document titled:
124441_AMD64_SpeculativeStoreBypassDisable_Whitepaper_final.pdf
gives priority of SPEC CTRL MSR over the VIRT SPEC CTRL MSR.
A copy of this document is available at
https://bugzilla.kernel.org/show_bug.cgi?id=199889
Anyhow, this means that on future AMD CPUs there will be _two_ ways to
deal with SSBD.
Backports commit a764f3f7197f4d7ad8fe8424269933de912224cb from qemu
OSPKE is not a static feature flag: it changes dynamically at
runtime depending on CR4, and it was never configurable: KVM
never returned OSPKE on GET_SUPPORTED_CPUID, and on TCG enables
it automatically if CR4_PKE_MASK is set.
Remove OSPKE from the feature name array so users don't try to
configure it manually.
Backports commit 9ccb9784b57804f5c74434ad6ccb66650a015ffc from qemu
OSXAVE is not a static feature flag: it changes dynamically at
runtime depending on CR4, and it was never configurable: KVM
never returned OSXSAVE on GET_SUPPORTED_CPUID, and it is not
included in TCG_EXT_FEATURES.
Remove OSXSAVE from the feature name array so users don't try to
configure it manually.
Backports commit f1a23522b03a569f13aad49294bb4c4b1a9500c7 from qemu
Add support for cpuid leaf CPUID_8000_001E. Build the config that closely
match the underlying hardware. Please refer to the Processor Programming
Reference (PPR) for AMD Family 17h Model for more details.
Backports commit ed78467a214595a63af7800a073a03ffe37cd7db from qemu
Add information for cpuid 0x8000001D leaf. Populate cache topology information
for different cache types (Data Cache, Instruction Cache, L2 and L3) supported
by 0x8000001D leaf. Please refer to the Processor Programming Reference (PPR)
for AMD Family 17h Model for more details.
Backports commit 8f4202fb1080f86958782b1fca0bf0279f67d136 from qemu
Always initialize CPUCaches structs with cache information, even
if legacy_cache=true. Use different CPUCaches struct for
CPUID[2], CPUID[4], and the AMD CPUID leaves.
This will simplify a lot the logic inside cpu_x86_cpuid()
Backports commit a9f27ea9adc8c695197bd08f2e938ef7b4183f07 from qemu
Do the cast to uintptr_t within the helper, so that the compiler
can type check the pointer argument. We can also do some more
sanity checking of the index argument.
Backports commit 07ea28b41830f946de3841b0ac61a3413679feb9 from qemu
AMD Zen expose the Intel equivalant to Speculative Store Bypass Disable
via the 0x80000008_EBX[25] CPUID feature bit.
This needs to be exposed to guest OS to allow them to protect
against CVE-2018-3639.
Backports commit 403503b162ffc33fb64cfefdf7b880acf41772cd from qemu
"Some AMD processors only support a non-architectural means of enabling
speculative store bypass disable (SSBD). To allow a simplified view of
this to a guest, an architectural definition has been created through a new
CPUID bit, 0x80000008_EBX[25], and a new MSR, 0xc001011f. With this, a
hypervisor can virtualize the existence of this definition and provide an
architectural method for using SSBD to a guest.
Add the new CPUID feature, the new MSR and update the existing SSBD
support to use this MSR when present." (from x86/speculation: Add virtualized
speculative store bypass disable support in Linux).
Backports commit cfeea0c021db6234c154dbc723730e81553924ff from qemu
New microcode introduces the "Speculative Store Bypass Disable"
CPUID feature bit. This needs to be exposed to guest OS to allow
them to protect against CVE-2018-3639.
Backports commit d19d1f965904a533998739698020ff4ee8a103da from qemu
The property legacy-cache will be used to control the cache information.
If user passes "-cpu legacy-cache" then older information will
be displayed even if the hardware supports new information. Otherwise
use the statically loaded cache definitions if available.
Renamed the previous cache structures to legacy_*. If there is any change in
the cache information, then it needs to be initialized in builtin_x86_defs.
Backports commit ab8f992e3e63e91be257e4e343d386dae7be4bcb from qemu
Instead of having a collection of macros that need to be used in
complex expressions to build CPUID data, define a CPUCacheInfo
struct that can hold information about a given cache. Helper
functions will take a CPUCacheInfo struct as input to encode
CPUID leaves for a cache.
This will help us ensure consistency between cache information
CPUID leaves, and make the existing inconsistencies in CPUID info
more visible.
Backports commit 7e3482f824809e1f6ffeb5bb8103ba27a7d1a52a from qemu
The CLDEMOTE instruction hints to hardware that the cache line that
contains the linear address should be moved("demoted") from
the cache(s) closest to the processor core to a level more distant
from the processor core. This may accelerate subsequent accesses
to the line by other cores in the same coherence domain,
especially if the line was written by the core that demotes the line.
Intel Snow Ridge has added new cpu feature, CLDEMOTE.
The new cpu feature needs to be exposed to guest VM.
The bit definition:
CPUID.(EAX=7,ECX=0):ECX[bit 25] CLDEMOTE
The release document ref below link:
https://software.intel.com/sites/default/files/managed/c5/15/\
architecture-instruction-set-extensions-programming-reference.pdf
Backports commit 0da0fb062841d0dcd8ba47e4a989d2e952cdf0ff from qemu
A new cpu model called "KnightsMill" is added to model Knights Mill
processors. Compared to "Skylake-Server" cpu model, the following
features are added:
avx512_4vnniw avx512_4fmaps avx512pf avx512er avx512_vpopcntdq
and the following features are removed:
pcid invpcid clflushopt avx512dq avx512bw clwb smap rtm mpx
xsavec xgetbv1 hle
Backports commit a18495159a35e9c5973d9aa0f612a97318bf684d from qemu
While at it, use int for both num_insns and max_insns to make
sure we have same-type comparisons.
Backports commit b542683d77b4f56cef0221b267c341616d87bce9 from qemu
In icount mode, instructions that access io memory spaces in the middle
of the translation block invoke TB recompilation. After recompilation,
such instructions become last in the TB and are allowed to access io
memory spaces.
When the code includes instruction like i386 'xchg eax, 0xffffd080'
which accesses APIC, QEMU goes into an infinite loop of the recompilation.
This instruction includes two memory accesses - one read and one write.
After the first access, APIC calls cpu_report_tpr_access, which restores
the CPU state to get the current eip. But cpu_restore_state_from_tb
resets the cpu->can_do_io flag which makes the second memory access invalid.
Therefore the second memory access causes a recompilation of the block.
Then these operations repeat again and again.
This patch moves resetting cpu->can_do_io flag from
cpu_restore_state_from_tb to cpu_loop_exit* functions.
It also adds a parameter for cpu_restore_state which controls restoring
icount. There is no need to restore icount when we only query CPU state
without breaking the TB. Restoring it in such cases leads to the
incorrect flow of the virtual time.
In most cases new parameter is true (icount should be recalculated).
But there are two cases in i386 and openrisc when the CPU state is only
queried without the need to break the TB. This patch fixes both of
these cases.
Backports commit afd46fcad2dceffda35c0586f5723c127b6e09d8 from qemu
In commit 7073fbada733c8d10992f00772c9b9299d740e9b, the `andn` instruction
was implemented via `tcg_gen_andc` but passes the operands in the wrong
order:
- X86 defines `andn dest,src1,src2` as: dest = ~src1 & src2
- TCG defines `andc dest,src1,src2` as: dest = src1 & ~src2
The following simple test shows the issue:
int main(void) {
uint32_t ret = 0;
__asm (
"mov $0xFF00, %%ecx\n"
"mov $0x0F0F, %%eax\n"
"andn %%ecx, %%eax, %%ecx\n"
"mov %%ecx, %0\n"
: "=r" (ret));
printf("%08X\n", ret);
return 0;
}
This patch fixes the problem by simply swapping the order of the two last
arguments in `tcg_gen_andc_tl`.
Backports commit 5cd10051c2e02b7a86eae49919d6c65a87dbea46 from qemu
Add an Error argument to cpu_exec_init() to let users collect the
error. This is in preparation to change the CPU enumeration logic
in cpu_exec_init(). With the new enumeration logic, cpu_exec_init()
can fail if cpu_index values corresponding to max_cpus have already
been handed out.
Since all current callers of cpu_exec_init() are from instance_init,
use error_abort Error argument to abort in case of an error.
Backports commit 5a790cc4b942e651fec7edc597c19b637fad5a76 from qemu
Intel processor trace should be disabled when
CPUID.(EAX=14H,ECX=0H).ECX.[bit31] is set.
Generated packets which contain IP payloads will have LIP
values when this bit is set, or IP payloads will have RIP
values.
Currently, The information of CPUID 14H is constant to make
live migration safty and this bit is always 0 in guest even
if host support LIP values.
Guest sees the bit is 0 will expect IP payloads with RIP
values, but the host CPU will generate IP payloads with
LIP values if this bit is set in HW.
To make sure the value of IP payloads correctly, Intel PT
should be disabled when bit[31] is set.
Backports relevant parts of commit c078ca968c6c7cb62781c1843d840cb0f5c72781 from qemu
cpu_init(cpu_model) were replaced by cpu_create(cpu_type) so
no users are left, remove it.
Backports commit 3f71e724e283233753f1b5b3d6a30948d3084636 from qemu
define default CPU type in generic way in pc_machine_class_init()
and let common machine code to handle cpu_model parsing
Patch also introduces TARGET_DEFAULT_CPU_TYPE define for 2 purposes:
* make foo_machine_class_init() look uniform on every target
* use define in [bsd|linux]-user targets to pick default
cpu type
Backports commit 311ca98d16bbb6a2a38b38ba898baa4a4d4ab9a7 from qemu
Considering that features are converted to global properties and
global properties are automatically applied to every new instance
of created CPU (at object_new() time), there is no point in
parsing cpu_model string every time a CPU created. So move
parsing outside CPU creation loop and do it only once.
Parsing also should be done before any CPU is created so that
features would affect the first CPU a well.
Backports commit 6aff24c6a61c6fec31e555c7748ba6085b7b2c06 from qemu
Considering that features are converted to global properties and
global properties are automatically applied to every new instance
of created CPU (at object_new() time), there is no point in
parsing cpu_model string every time a CPU created. So move
parsing outside CPU creation loop and do it only once.
Parsing also should be done before any CPU is created so that
features would affect the first CPU a well.
Backports commit 6aff24c6a61c6fec31e555c7748ba6085b7b2c06 from qemu
Currently CPUClass->parse_features() is used to parse -cpu
features string and set properties on created CPU instances.
But considering that features specified by -cpu apply to every
created CPU instance, it doesn't make sense to parse the same
features string for every CPU created. It also makes every target
that cares about parsing features string explicitly call
CPUClass->parse_features() parser, which gets in a way if we
consider using generic device_add for CPU hotplug as device_add
has not a clue about CPU specific hooks.
Turns out we can use global properties mechanism to set
properties on every created CPU instance for a given type. That
way it's possible to convert CPU features into a set of global
properties for CPU type specified by -cpu cpu_model and common
Device.device_post_init() will apply them to CPU of given type
automatically regardless whether it's manually created CPU or CPU
created with help of device_add.
Backports commits 62a48a2a5798425997152dea3fc48708f9116c04 and
f313369fdb78f849ecbbd8e5d88f01ddf38786c8 from qemu
it will be used for providing to cpu name resolving class for
parsing cpu model for system and user emulation code.
Along with change add target to null-machine tests, so
that when switch to CPU_RESOLVING_TYPE happens,
it would ensure that null-machine usecase still works.
Backports commit 0dacec874fa3b3fd34b0d0670fa257efdcbbebd0 from qemu
Backports commits 2994fd96d986578a342f2342501b4ad30f6d0a85,
701e3c78ce45fa630ffc6826c4b9a4218954bc7f, and
d1853231c60d16af78cf4d1608d043614bfbac0b from qemuu
This function needs to be converted to QOM hook and virtualised for
multi-arch. This rename interferes, as cpu-qom will not have access
to the renaming causing name divergence. This rename doesn't really do
anything anyway so just delete it.
Backports commit 8642c1b81e0418df066a7960a7426d85a923a253 from qemu
Move vcpu's associated numa_node field out of generic CPUState
into inherited classes that actually care about cpu<->numa mapping,
i.e: ARMCPU, PowerPCCPU, X86CPU.
Backports relevant parts of commit 15f8b14228b856850df3fa5ba999ad96521f2208 from qemu
Add Intel Processor Trace related definition. It also add
corresponding part to kvm_get/set_msr and vmstate.
Backports commit b77146e9a129bcdb60edc23639211679ae846a92 from qemu
Expose Intel Processor Trace feature to guest.
To make Intel PT live migration safe and get same CPUID information
with same CPU model on diffrent host. CPUID[14] is constant in this
patch. Intel PT use EPT is first supported in IceLake, the CPUID[14]
get on this machine as default value. Intel PT would be disabled
if any machine don't support this minial feature list.
Backports commit e37a5c7fa459558b5020588994707fe3fdd6616e from qemu
Add KVM_HINTS_DEDICATED performance hint, guest checks this feature bit
to determine if they run on dedicated vCPUs, allowing optimizations such
as usage of qspinlocks.
Backports commit be7773268d98176489483a315d3e2323cb0615b9 from qemu
This MSR returns the number of #SMIs that occurred on
CPU since boot.
KVM commit 52797bf9a875 ("KVM: x86: Add emulation of MSR_SMI_COUNT")
introduced support for emulating this MSR.
This commit adds support for QEMU to save/load this
MSR for migration purposes.
Backports relevant parts of commit e13713db5b609d9a83c9cfc8ba389d4215d4ba29 from qemu
When SEV is enabled, CPUID 0x8000_001F should provide additional
information regarding the feature (such as which page table bit is used
to mark the pages as encrypted etc).
The details for memory encryption CPUID is available in AMD APM
(https://support.amd.com/TechDocs/24594.pdf) Section E.4.17
Backports relevant parts of commit 6cb8f2a663a47c6e0da17fc4fb9e06abfda2bd48 from qemu
Now that we have a DEFINE_PC_MACHINE helper macro that just requires an
initialization function, it is trivial to convert them to register a QOM
machine class directly, instead of using QEMUMachine.
Backports commit 865906f7fdadd2732441ab158787f81f6a212bfe from qemu
Clang 3.9 passes the CONFIG_AVX2_OPT configure test. However, the
supplied <cpuid.h> does not contain the bit_AVX2 define that we use
when detecting whether the routine can be enabled.
Introduce a qemu-specific header that uses the compiler's definition
of __cpuid et al, but supplies any missing bit_* definitions needed.
This avoids introducing any extra ifdefs to util/bufferiszero.c, and
allows quite a few to be removed from tcg/i386/tcg-target.inc.c.
Backports commit 5dd8990841a9e331d9d4838a116291698208cbb6 from qemu
As cpu.h is another typically widely included file which doesn't need
full access to the softfloat API we can remove the includes from here
as well. Where they do need types it's typically for float_status and
the rounding modes so we move that to softfloat-types.h as well.
As a result of not having softfloat in every cpu.h call we now need to
add it to various helpers that do need the full softfloat.h
definitions.
Backports commit 24f91e81b65fcdd0552d1f0fcb0ea7cfe3829c19 from qemu
This cleanup makes the number of objects depending on qapi/qmp/qdict.h
drop from 4550 (out of 4743) to 368 in my "build everything" tree.
For qapi/qmp/qobject.h, the number drops from 4552 to 390.
While there, separate #include from file comment with a blank line.
Backports commit 452fcdbc49c59884c8c284268d64baa24fea11e1 from qemu
This cleanup makes the number of objects depending on qapi/error.h
drop from 1910 (out of 4743) to 1612 in my "build everything" tree.
While there, separate #include from file comment with a blank line,
and drop a useless comment on why qemu/osdep.h is included first.
Backports commit e688df6bc4549f28534cdb001f168b8caae55b0c from qemu
The MC68040 MMU provides the size of the access that
triggers the page fault.
This size is set in the Special Status Word which
is written in the stack frame of the access fault
exception.
So we need the size in m68k_cpu_unassigned_access() and
m68k_cpu_handle_mmu_fault().
To be able to do that, this patch modifies the prototype of
handle_mmu_fault handler, tlb_fill() and probe_write().
do_unassigned_access() already includes a size parameter.
This patch also updates handle_mmu_fault handlers and
tlb_fill() of all targets (only parameter, no code change).
Backports commit 98670d47cd8d63a529ff230fd39ddaa186156f8c from qemu
cpu_restore_state officially supports being passed an address it can't
resolve the state for. As a result the checks in the helpers are
superfluous and can be removed. This makes the code consistent with
other users of cpu_restore_state.
Of course this does nothing to address what to do if cpu_restore_state
can't resolve the state but so far it seems this is handled elsewhere.
The change was made with included coccinelle script.
Backports commit 65255e8efdd5fca602bcc4ff61a879939ff75f4f from qemu
Normally we create an address space for that CPU and pass that address
space into the function. Let's just do it inside to unify address space
creations. It'll simplify my next patch to rename those address spaces.
Backports commit 80ceb07a83375e3a0091591f96bd47bce2f640ce from qemu
These gcc warnings are fixed:
target/i386/translate.c:4461:12: warning:
variable 'prefixes' might be clobbered by 'longjmp' or 'vfork' [-Wclobbered]
target/i386/translate.c:4466:9: warning:
variable 'rex_w' might be clobbered by 'longjmp' or 'vfork' [-Wclobbered]
target/i386/translate.c:4466:16: warning:
variable 'rex_r' might be clobbered by 'longjmp' or 'vfork' [-Wclobbered]
Tested with x86_64-w64-mingw32-gcc from Debian stretch.
Backports commit a4926d99129a1d8072fc4681cd4efdb214f65ed4 from qemu
Intel IceLake cpu has added new cpu features,AVX512_VBMI2/GFNI/
VAES/VPCLMULQDQ/AVX512_VNNI/AVX512_BITALG. Those new cpu features
need expose to guest VM.
The bit definition:
CPUID.(EAX=7,ECX=0):ECX[bit 06] AVX512_VBMI2
CPUID.(EAX=7,ECX=0):ECX[bit 08] GFNI
CPUID.(EAX=7,ECX=0):ECX[bit 09] VAES
CPUID.(EAX=7,ECX=0):ECX[bit 10] VPCLMULQDQ
CPUID.(EAX=7,ECX=0):ECX[bit 11] AVX512_VNNI
CPUID.(EAX=7,ECX=0):ECX[bit 12] AVX512_BITALG
The release document ref below link:
https://software.intel.com/sites/default/files/managed/c5/15/\
architecture-instruction-set-extensions-programming-reference.pdf
Backports commit aff9e6e46a343e1404498be4edd03db1112f0950 from qemu
EPYC-IBPB is a copy of the EPYC CPU model with
just CPUID_8000_0008_EBX_IBPB added.
Backports commit 8ebfafa796ca0cb2b035a7f06f836a675d8b48be from qemu
The new MSR IA32_SPEC_CTRL MSR was introduced by a recent Intel
microcode updated and can be used by OSes to mitigate
CVE-2017-5715. Unfortunately we can't change the existing CPU
models without breaking existing setups, so users need to
explicitly update their VM configuration to use the new *-IBRS
CPU model if they want to expose IBRS to guests.
The new CPU models are simple copies of the existing CPU models,
with just CPUID_7_0_EDX_SPEC_CTRL added and model_id updated.
Backports commit 61efbbf869293f1deb9ee39d44bd4e635de59fa7 from qemu
Add the new feature word and the "ibpb" feature flag.
Based on a patch by Paolo Bonzini.
Backports commit 1ade973f5202404e772aae7b1acd331270d246dc from qemu
It is valid to have a 48-character model ID on CPUID, however the
definition of X86CPUDefinition::model_id is char[48], which can
make the compiler drop the null terminator from the string.
If a CPU model happens to have 48 bytes on model_id, "-cpu help"
will print garbage and the object_property_set_str() call at
x86_cpu_load_def() will read data outside the model_id array.
We could increase the array size to 49, but this would mean the
compiler would not issue a warning if a 49-char string is used by
mistake for model_id.
To make things simpler, simply change model_id to be const char*,
and validate the string length using an assert() on
x86_register_cpudef_type().
Backports commit 4b220d88ba76fb2623ce4b8ba1f1eea66b82144e from qemu
In commit e3af7c788b73a6495eb9d94992ef11f6ad6f3c56 we
replaced direct calls to to cpu_ld*_code() with calls
to the x86_ld*_code() wrappers which incorporate an
advance of s->pc. Unfortunately we didn't notice that
in one place the old code was deliberately not incrementing
s->pc:
@@ -4501,7 +4528,7 @@ static target_ulong disas_insn(DisasContext *s, CPUState *cpu)
static const int pp_prefix[4] = {
0, PREFIX_DATA, PREFIX_REPZ, PREFIX_REPNZ
};
- int vex3, vex2 = cpu_ldub_code(env, s->pc);
+ int vex3, vex2 = x86_ldub_code(env, s);
if (!CODE64(s) && (vex2 & 0xc0) != 0xc0) {
/* 4.1.4.6: In 32-bit mode, bits [7:6] must be 11b,
This meant we were mishandling this set of instructions.
Remove the manual advance of s->pc for the "is VEX" case
(which is now done by x86_ldub_code()) and instead rewind
PC in the case where we decide that this isn't really VEX.
Backports commit 817a9fcba8043faa467929e7b0193df6bdc92211 from qemu
When we used structures for TCGv_*, we needed a macro in order to
perform a comparison. Now that we use pointers, this is just clutter
Backports commit 11f4e8f8bfaa2caaab24bef6bbbb8a0205015119 from qemu
Rather than have a separate buffer of 10*max_ops entries,
give each opcode 10 entries. The result is actually a bit
smaller and should have slightly more cache locality.
Backports commit 75e8b9b7aa0b95a761b9add7e2f09248b101a392 from qemu
Besides being more correct, arbitrarily long instruction allow the
generation of a translation block that spans three pages. This
confuses the generator and even allows ring 3 code to poison the
translation block cache and inject code into other processes that are
in guest ring 3.
This is an improved (and more invasive) fix for commit 30663fd ("tcg/i386:
Check the size of instruction being translated", 2017-03-24). In addition
to being more precise (and generating the right exception, which is #GP
rather than #UD), it distinguishes better between page faults and too long
instructions, as shown by this test case:
int main()
{
char *x = mmap(NULL, 8192, PROT_READ|PROT_WRITE|PROT_EXEC,
MAP_PRIVATE|MAP_ANON, -1, 0);
memset(x, 0x66, 4096);
x[4096] = 0x90;
x[4097] = 0xc3;
char *i = x + 4096 - 15;
mprotect(x + 4096, 4096, PROT_READ|PROT_WRITE);
((void(*)(void)) i) ();
}
... which produces a #GP without the mprotect, and a #PF with it.
Backports commit b066c5375737ad0d630196dab2a2b329515a1d00 from qemu
These take care of advancing s->pc, and will provide a unified point
where to check for the 15-byte instruction length limit.
Backports commit e3af7c788b73a6495eb9d94992ef11f6ad6f3c56 from qemu
It is unlikely that we will ever want to call this helper passing
an argument other than the current PC. So just remove the argument,
and use the pc we already get from cpu_get_tb_cpu_state.
This change paves the way to having a common "tb_lookup" function.
Backports commit 7f11636dbee89b0e4d03e9e2b96e14649a7db778 from qemu
It looks like there was a transcription error when writing this code
initially. The code previously only decoded src or dst of rax. This
resolves
https://bugs.launchpad.net/qemu/+bug/1719984.
Backports commit e0dd5fd41a1a38766009f442967fab700d2d0550 from qemu
Starting with Windows Server 2012 and Windows 8, if
CPUID.40000005.EAX contains a value of -1, Windows assumes specific
limit to the number of VPs. In this case, Windows Server 2012
guest VMs may use more than 64 VPs, up to the maximum supported
number of processors applicable to the specific Windows
version being used.
https://docs.microsoft.com/en-us/virtualization/hyper-v-on-windows/reference/tlfs
For compatibility, Let's introduce a new property for X86CPU,
named "x-hv-max-vps" as Eduardo's suggestion, and set it
to 0x40 before machine 2.10.
(The "x-" prefix indicates that the property is not supposed to
be a stable user interface.)
Backports relevant parts of commit 6c69dfb67e84747cf071958594d939e845dfcc0c from qemu
The SSE4.1 phminposuw instruction finds the minimum 16-bit element in
the source vector, putting the value of that element in the low 16
bits of the destination vector, the index of that element in the next
three bits and zeroing the rest of the destination. The helper for
this operation fills the destination from high to low, meaning that
when the source and destination are the same register, the minimum
source element can be overwritten before it is copied to the
destination. This patch fixes it to fill the destination from low to
high instead, so the minimum source element is always copied first.
This fixes one gcc test failure in my GCC 6-based testing (and so
concludes the present sequence of patches, as I don't have any further
gcc test failures left in that testing that I attribute to QEMU bugs).
Backports commit aa406feadfc5b095ca147ec56d6187c64be015a7 from qemu
One of the cases of the SSE4.2 pcmpestri / pcmpestrm / pcmpistri /
pcmpistrm instructions does a substring search. The implementation of
this case in the pcmpxstrx helper is incorrect. The operation in this
case is a search for a string (argument d to the helper) in another
string (argument s to the helper); if a copy of d at a particular
position would run off the end of s, the resulting output bit should
be 0 whether or not the strings match in the region where they
overlap, but the QEMU implementation was wrongly comparing only up to
the point where s ends and counting it as a match if an initial
segment of d matched a terminal segment of s. Here, "run off the end
of s" means that some byte of d would overlap some byte outside of s;
thus, if d has zero length, it is considered to match everywhere,
including after the end of s. This patch fixes the implementation to
correspond with the proper instruction semantics. This fixes four gcc
test failures in my GCC 6-based testing.
Backports commit ae35eea7e4a9f21dd147406dfbcd0c4c6aaf2a60 from qemu
The SSE4.1 packusdw instruction combines source and destination
vectors of signed 32-bit integers into a single vector of unsigned
16-bit integers, with unsigned saturation. When the source and
destination are the same register, this means each 32-bit element of
that register is used twice as an input, to produce two of the 16-bit
output elements, and so if the operation is carried out
element-by-element in-place, no matter what the order in which it is
applied to the elements, the first element's operation will overwrite
some future input. The helper for packssdw avoids this issue by
computing the result in a local temporary and copying it to the
destination at the end; this patch fixes the packusdw helper to do
likewise. This fixes three gcc test failures in my GCC 6-based
testing.
Backports commit 80e19606215d4df370dfe8fe21c558a129f00f0b from qemu
It turns out that my recent fix to set rip_offset when emulating some
SSE4.1 instructions needs generalizing to cover a wider class of
instructions. Specifically, every instruction in the sse_op_table7
table, coming from various instruction set extensions, has an 8-bit
immediate operand that comes after any memory operand, and so needs
rip_offset set for correctness if there is a memory operand that is
rip-relative, and my patch only set it for a subset of those
instructions. This patch moves the rip_offset setting to cover the
wider class of instructions, so fixing 9 further gcc testsuite
failures in my GCC 6-based testing. (I do not know whether there
might be still further classes of instructions missing this setting.)
Backports commit c6a8242915328cda0df0fbc0803da3448137e614 from qemu
The SSE4.1 pmovsx* and pmovzx* instructions take packed 1-byte, 2-byte
or 4-byte inputs and sign-extend or zero-extend them to a wider vector
output. The associated helpers for these instructions do the
extension on each element in turn, starting with the lowest. If the
input and output are the same register, this means that all the input
elements after the first have been overwritten before they are read.
This patch makes the helpers extend starting with the highest element,
not the lowest, to avoid such overwriting. This fixes many GCC test
failures (161 in the gcc testsuite in my GCC 6-based testing) when
testing with a default CPU setting enabling those instructions.
Backports commit c6a56c8e990b213a1638af2d34352771d5fa4d9c from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit e0d110d943891b719de7ca075fc17fa8ea5749b8 from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit 47e981b42553f00110024c33897354f9014e83e9 from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit 2c2f8cacd8cf4f67d6f1384b19d38f9a0a25878b from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit e6b41ec37f0a9742374dfdb90e662745969cd7ea from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit e6b41ec37f0a9742374dfdb90e662745969cd7ea from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit 9d75f52b34053066b8e8fc37610d5f300d67538b from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit 9761d39b09c4beb1340bf3074be3d3e0a5d453a4 from qemu
Incrementally paves the way towards using the generic instruction translation
loop.
Backports commit 6cf147aa299e49f7794858609a1e8ef19f81c007 from qemu
Used later. An enum makes expected values explicit and
bounds the value space of switches.
Backports commit 77fc6f5e28667634916f114ae04c6029cd7b9c45 from qemu
This target is not sophisticated in its use of cleanups at the
end of the translation loop. For the most part, any condition
that exits the TB is dealt with by emitting the exiting opcode
right then and there. Therefore the only is_jmp indicator that
is needed is DISAS_NORETURN.
For two stack segment modifying cases, we have not yet exited
the TB (therefore DISAS_NORETURN feels wrong), but intend to exit.
The caller of gen_movl_seg_T0 currently checks for any non-zero
value, therefore DISAS_TOO_MANY seems acceptable for that usage.
Backports commit 1e39d97af086d525cd0408eaa5d19783ea165906 from qemu
Add a new base CPU model called 'EPYC' to model processors from AMD EPYC
family (which includes EPYC 76xx,75xx,74xx, 73xx and 72xx).
The following features bits have been added/removed compare to Opteron_G5
Added: monitor, movbe, rdrand, mmxext, ffxsr, rdtscp, cr8legacy, osvw,
fsgsbase, bmi1, avx2, smep, bmi2, rdseed, adx, smap, clfshopt, sha
xsaveopt, xsavec, xgetbv1, arat
Removed: xop, fma4, tbm
Backports commit 2e2efc7dbe2b0adc1200b5aa286cdbed729f6751 from qemu
When emulating various SSE4.1 instructions such as pinsrd, the address
of a memory operand is computed without allowing for the 8-bit
immediate operand located after the memory operand, meaning that the
memory operand uses the wrong address in the case where it is
rip-relative. This patch adds the required rip_offset setting for
those instructions, so fixing some GCC test failures (13 in the gcc
testsuite in my GCC 6-based testing) when testing with a default CPU
setting enabling those instructions.
Backports commit ab6ab3e9972a49a359f59895a88bed311472ca97 from qemu
Introduce Skylake-Server cpu mode which inherits the features from
Skylake-Client and supports some additional features that are: AVX512,
CLWB and PGPE1GB.
Backports commit 53f9a6f45fb214540cb40af45efc11ac40ac454c from qemu
Currently when running KVM, we expose "KVMKVMKVM\0\0\0" in
the 0x40000000 CPUID leaf. Other hypervisors (VMWare,
HyperV, Xen, BHyve) all do the same thing, which leaves
TCG as the odd one out.
The CPUID signature is used by software to detect which
virtual environment they are running in and (potentially)
change behaviour in certain ways. For example, systemd
supports a ConditionVirtualization= setting in unit files.
The virt-what command can also report the virt type it is
running on
Currently both these apps have to resort to custom hacks
like looking for 'fw-cfg' entry in the /proc/device-tree
file to identify TCG.
This change thus proposes a signature "TCGTCGTCGTCG" to be
reported when running under TCG.
To hide this, the -cpu option tcg-cpuid=off can be used.
Backports commits 4ed3d478c63dc65a02eba774c35116618ea5ff10 and 1ce36bfe6424243082d3d7c2330e1a0a4ff72a43 from qemu
Move cpu_get_fp80()/cpu_set_fp80() from fpu_helper.c to
machine.c because fpu_helper.c will be disabled if tcg is
disabled in the build.
Backports commit db573d2cf7ae6b5a4fc324be6f55e078fc218464 from qemu.
In unicorn's case, they can be moved into unicorn.c
Move cpu_sync_bndcs_hflags() function from mpx_helper.c
to helper.c because mpx_helper.c need be disabled when
tcg is disabled.
Backports commit ab0a19d4f08d924e052eb369420d264240872f8a from qemu
Move the handling of conforming code segments before the handling
of stack switch.
Because dpl == cpl after the new "if", it's now unnecessary to check
the C bit when testing dpl < cpl. Furthermore, dpl > cpl is checked
slightly above the modified code, so the final "else" is unreachable
and we can remove it.
Backports commit 1110bfe6f5600017258fa6578f9c17ec25b32277 from qemu
In do_interrupt64(), when interrupt stack table(ist) is enabled
and the the target code segment is conforming(e2 & DESC_C_MASK), the
old implementation always set new CPL to 0, and SS.RPL to 0.
This is incorrect for when CPL3 code access a CPL0 conforming code
segment, the CPL should remain unchanged. Otherwise higher privileged
code can be compromised.
The patch fix this for always set dpl = cpl when the target code segment
is conforming, and modify the last parameter `flags`, which contains
correct new CPL, in cpu_x86_load_seg_cache().
Backports commit e95e9b88ba5f4a6c17f4d0c3a3a6bf3f648bb328 from qemu
Running Windows with icount causes a crash in instruction of write cr.
This patch fixes it.
Reading and writing cr cause an icount read because there are called
cpu_get_apic_tpr and cpu_set_apic_tpr functions. So, there is need
gen_io_start()/gen_io_end() calls.
Backports commit 5b003a40bb1ab14d0398e91f03393d3c6b9577cd from qemu
This speeds up SMM switches. Later on it may remove the need to take
the BQL, and it may also allow to reuse code between TCG and KVM.
Backports commit f8c45c6550b9ff1e1f0b92709ff3213a79870879 from qemu
Instead of unconditionally exiting to the exec loop, use the
gen_jr helper to jump to the target if it is valid.
Perf impact: see next commit's log.
Backports commit fe62089563ffc6a42f16ff28a6b6be34d2697766 from qemu
The existing code for "host" and "max" CPU models overrides every
single feature in the CPU object at realize time, even the ones
that were explicitly enabled or disabled by the user using
"feat=on" or "feat=off", while features set using +feat/-feat are
kept.
This means "-cpu host,+invtsc" works as expected, while
"-cpu host,invtsc=on" doesn't.
This was a known bug, already documented in a comment inside
x86_cpu_expand_features(). What makes this bug worse now is that
libvirt 3.0.0 and newer now use "feat=on|off" instead of
+feat/-feat when it detects a QEMU version that supports it (see
libvirt commit d47db7b16dd5422c7e487c8c8ee5b181a2f9cd66).
Change the feature property getter/setter to set a
env->user_features field, to keep track of features that were
explicitly changed using QOM properties. Then make the
max_features code not override user features when handling "-cpu
host" and "-cpu max".
This will also allow us to remove the plus_features/minus_features
hack in the future, but I plan to do that after 2.9.0 is
released.
Backports commit d4a606b38b5d4b3689b86cc1575908e82179ecfb from qemu
glibc blacklists TSX on Haswell CPUs with model==60 and
stepping < 4. To make the Haswell CPU model more useful, make
those guests actually use TSX by changing CPU stepping to 4.
References:
* glibc commit 2702856bf45c82cf8e69f2064f5aa15c0ceb6359
https://sourceware.org/git/?p=glibc.git;a=commit;h=2702856bf45c82cf8e69f2064f5aa15c0ceb6359
Backports commit ec56a4a7b07e2943f49da273a31e3195083b1f2e from qemu
Helper function for code that needs to check the host CPU
vendor/family/model/stepping values.
Backports commit 20271d484069f154fb262507e63adc3a37e885d2 from qemu
Paths through the softmmu code during code generation now need to be audited
to check for double locking of tb_lock. In particular, VMEXIT can take tb_lock
through cpu_vmexit -> cpu_x86_update_cr4 -> tlb_flush.
To avoid this, split VMEXIT delivery in two parts, similar to what is done with
exceptions. cpu_vmexit only records the VMEXIT exit code and information, and
cc->do_interrupt can then deliver it when it is safe to take the lock.
Backports commit 10cde894b63146139f981857e4eedf756fa53dcb from qemu
This suppresses the incorrect warning when forcing MTTCG for x86
guests on x86 hosts. A future patch will still warn when
TARGET_SUPPORT_MTTCG hasn't been defined for the guest (which is still
pending for x86).
Backports commit 72c1701f62e8d44eb24a0583a958edc280105455 from qemu
CPU runnability checks and CPU model expansion have slightly
different requirements. Document the steps involved in loading a
CPU model and realizing a CPU, so their requirements and purpose
are clearly defined.
This patch doesn't change any implementation. It just add
comments, rename the x86_cpu_load_features() function for clarity
(so it won't be confused with x86_cpu_load_def()), and move
x86_cpu_filter_features() closer to it.
Backports commit b8d834a00fa3ed4dad7d371e1a00938a126a54a0 from qemu
This makes qemu_strtosz(), qemu_strtosz_mebi() and
qemu_strtosz_metric() similar to qemu_strtoi64(), except negative
values are rejected.
Backports commit f17fd4fdf0df3d2f3444399d04c38d22b9a3e1b7 from qemu
Change the qemu_strtosz() & friends to return -EINVAL when @endptr is
null and the conversion doesn't consume the string completely.
Matches how qemu_strtol() & friends work.
Only test_qemu_strtosz_simple() passes a null @endptr. No functional
change there, because its conversion consumes the string.
Simplify callers that use @endptr only to fail when it doesn't point
to '\0' to pass a null @endptr instead.
Backports commit 4fcdf65ae2c00ae69f7625f26ed41f37d77b403c from qemu
To parse numbers with metric suffixes, we use
qemu_strtosz_suffix_unit(nptr, &eptr, QEMU_STRTOSZ_DEFSUFFIX_B, 1000)
Capture this in a new function for legibility:
qemu_strtosz_metric(nptr, &eptr)
Replace test_qemu_strtosz_suffix_unit() by test_qemu_strtosz_metric().
Rename qemu_strtosz_suffix_unit() to do_strtosz() and give it internal
linkage.
Backports commit d2734d2629266006b0413433778474d5801c60be from qemu
When CPU vendor is set to AMD, the AMD feature alias bits on
CPUID[0x80000001].EDX are already automatically copied from CPUID[1].EDX
on x86_cpu_realizefn(). When CPU vendor is Intel, those bits are
reserved and should be zero. On either case, those bits shouldn't be set
in the CPU model table.
Commit 726a8ff68677d8d5fba17eb0ffb85076bfb598dc removed those
bits from most CPU models, but the Opteron_* entries still have
them. Remove the alias bits from Opteron_* too.
Add an assert() to x86_register_cpudef_type() to ensure we don't
make the same mistake again.
Backports commit 2a923a293df95334fa22634016efdd138f49da7f from qemu
AVX512_VPOPCNTDQ: Vector POPCNT instructions for word and qwords.
variable precision.
Backports commit f77543772dcd38fa438470d9b80bafbd3a3ebbd7 from qemu
We've currently got 18 architectures in QEMU, and thus 18 target-xxx
folders in the root folder of the QEMU source tree. More architectures
(e.g. RISC-V, AVR) are likely to be included soon, too, so the main
folder of the QEMU sources slowly gets quite overcrowded with the
target-xxx folders.
To disburden the main folder a little bit, let's move the target-xxx
folders into a dedicated target/ folder, so that target-xxx/ simply
becomes target/xxx/ instead.
Backports commit fcf5ef2ab52c621a4617ebbef36bf43b4003f4c0 from qemu
