We have checks in both xen/compiler.h, and Config.mk. Both are incomplete.
The check in Config.mk sees $(CC) in system and cross-compiler form, so cannot
express anything more than the global baseline. Change it to simply 5.1.
In xen/compiler.h, rewrite the expression for clarity/brevity.
Include a GCC 12.2 check for RISCV, and include a Clang 11 baseline check.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
These are all loops over a scalar value, and don't need to call general bitop
helpers behind the scenes.
Clamp data to the width of the access in dispatch_mmio_write(), rather than
doing so in every handler.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
The bitmap_for_each() expression only inspects the bottom 8 bits of targets.
Change it's type to uint8_t and use for_each_set_bit() which is more efficient
over scalars.
GICD_SGI_TARGET_LIST_MASK is 2 bits wide. Two cases discard the prior
calculation of targets, and one case exits early.
Therefore, move the GICD_SGI_TARGET_MASK calculation into the only case which
wants it, and use MASK_EXTR() to simplify the expression.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Michal Orzel <michal.orzel@amd.com>
The switch() statement is over bits 24:25 (unshifted) of the guest provided
value. This makes case 0x3: dead, and not an implementation of the 4th
possible state.
A guest which writes (0x3 << 24) | (0xff << 16) to this register will skip the
early exit, then enter bitmap_for_each() with targets not bound by nr_vcpus.
If the guest has fewer than 8 vCPUs, bitmap_for_each() will read off the end
of d->vcpu[] and use the resulting vcpu pointer to ultimately derive irq, and
perform out-of-bounds writes.
Fix this by changing case 0x3 to default.
Fixes: 08c688ca64 ("ARM: new VGIC: Add SGIR register handler")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Volodymyr Babchuk <volodymyr_babchuk@epam.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Manually break the false dependency for the benefit of cases such as
bitmap_weight() which is a reasonable hotpath.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
I was mistaken about when ASM_CALL_CONSTRAINT is applicable. It is not
applicable for plain pushes/pops, so remove it from the flags logic.
Clarify the description of ASM_CALL_CONSTRAINT to be explicit about unwinding
using framepointers.
Fixes: 0754534b8a ("x86/elf: Improve code generation in elf_core_save_regs()")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Petr reports:
(XEN) MMIO emulation failed (1): d12v1 64bit @ 0010:fffff8057ba7dfbf -> 45 0f 20 c2 ...
during introspection.
This is MOV %cr8, which is wired up for hvm_mov_{to,from}_cr(); the VMExit
fastpaths, but not for the full emulation slowpaths.
Xen's handling of %cr8 turns out to be quite wrong. At a minimum, we need
storage for %cr8 separate to APIC_TPR, and to alter intercepts based on
whether the vLAPIC is enabled or not. But that's more work than there is time
for in the short term, so make a stopgap fix.
Extend hvmemul_{read,write}_cr() with %cr8 cases. Unlike hvm_mov_to_cr(),
hardware hasn't filtered out invalid values (#GP checks are ahead of
intercepts), so introduce X86_CR8_VALID_MASK.
Reported-by: Petr Beneš <w1benny@gmail.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
In hvmemul_read_cr(), make the TRACE()/X86EMUL_OKAY path common in preparation
for adding a %cr8 case. Use a local 'val' variable instead of always
operating on a deferenced pointer.
In both, calculate curr once.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Drop CentOS 7 entirely. It's way to old now.
Ubuntu 22.04 is the oldest Ubuntu with a suitable version of Clang, so swap
the 16.04 clang builds for 22.04.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
Reviewed-by: Anthony PERARD <anthony.perard@vates.tech>
Something I overlooked when last cleaning up exception handling is that a TSS
is not necessary if IST isn't configured, and IST isn't necessary until we're
running guest code.
Introduce early_traps_init(), and rearrange the existing logic between this
and traps_init() later on boot, to allow defering TSS and IST setup.
In early_traps_init(), load the IDT and invalidate TR/LDTR; this sufficient
system-table setup to make exception handling work. The setup of the BSPs
per-cpu variables stay early too; they're used on certain error paths.
Move load_system_tables() later into traps_init(). Note that it already
contains enable_each_ist(), so this call is simply dropped.
This removes some complexity prior to having exception support, and lays the
groundwork to not even allocate a TSS when using FRED.
No practical change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
... and rename to traps_init() for consistency. Move the declaration from
asm/system.h into asm/traps.h.
This also involves moving init_ler() and variables. Move the declaration of
ler_msr from asm/msr.h to asm/traps.h.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Move the declaration from asm/system.h into asm/traps.h.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
cpu_init() doesn't particularly belong in trap_init(). This brings the BSP
more in line with the APs.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
The expression for one parameter of find_memory() is already complicated and
about to become moreso. Break it out into a new variable, and express it in
an easier-to-follow way.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>
This needs to be ahead of everything. Right now, it is after xen/init.h being
included for -DINIT_SECTIONS_ONLY
# 1 "./include/xen/compiler.h" 1
# 83 "./include/xen/compiler.h"
# 1 "./include/xen/init.h" 1
# 62 "./include/xen/init.h"
typedef int (*initcall_t)(void);
typedef void (*exitcall_t)(void);
# 72 "./include/xen/init.h"
void do_presmp_initcalls(void);
void do_initcalls(void);
# 84 "./include/xen/compiler.h" 2
# 122 "./include/xen/compiler.h"
#pragma GCC visibility push(hidden)
Fixes: 84c4461b7d ("Force out-of-line instances of inline functions into .init.text in init-only code")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
GCC 4.1.2 is from 2007, and Binutils 2.16 is a similar vintage. Clang 3.5 is
from 2014. Supporting toolchains this old is a massive development and
testing burden.
Set a minimum baseline of GCC 5.1 across the board, along with Binutils 2.25
which is the same age. These were chosen *3 years ago* as Linux's minimum
requirements because even back then, they were ubiquitous in distros. Choose
Clang/LLVM 11 as a baseline for similar reasons; the Linux commit making this
change two years ago cites a laudry list of code generation bugs.
This will allow us to retire a lot of compatiblity logic, and start using new
features previously unavailable because of no viable compatibility option.
Merge the ARM 32bit and 64bit sections now they're the same.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Julien Grall <jgrall@amazon.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
The current CI failures turn out to be a latent bug triggered by a narrow set
of properties of the initrd and the host memory map, which CI encountered by
chance.
One step during boot involves constructing directmap mappings for modules.
With some probing at the point of creation, it is observed that there's a 4k
mapping missing towards the end of the initrd.
(XEN) === Mapped Mod1 [0000000394001000, 00000003be1ff6dc] to Directmap
(XEN) Probing paddr 394001000, va ffff830394001000
(XEN) Probing paddr 3be1ff6db, va ffff8303be1ff6db
(XEN) Probing paddr 3bdffffff, va ffff8303bdffffff
(XEN) Probing paddr 3be001000, va ffff8303be001000
(XEN) Probing paddr 3be000000, va ffff8303be000000
(XEN) Early fatal page fault at e008:ffff82d04032014c (cr2=ffff8303be000000, ec=0000)
The conditions for this bug appear to be map_pages_to_xen() call with a start
address of exactly 4k beyond a 2M boundary, some number of full 2M pages, then
a tail needing 4k pages.
Anyway, the condition for spotting superpage boundaries in map_pages_to_xen()
is wrong. The IS_ALIGNED() macro expects a power of two for the alignment
argument, and subtracts 1 itself.
Fixing this causes the failing case to now boot.
Fixes: 97fb6fcf26 ("x86/mm: introduce helpers to detect super page alignment")
Debugged-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
... which is more consise than the opencoded form, and more efficient when
compiled.
Furthermore, now that find_{first,next}_bit() are no longer in use, the
seg_reg_{accessed,dirty} fields aren't forced to be unsigned long, although
they do need to remain unsigned int because of __set_bit() elsewhere.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Regular data access into the trampoline is via the directmap.
As now discussed quite extensively in asm/trampoline.h, the trampoline is
arranged so that only the AP and S3 paths need an identity mapping, and that
they fit within a single page.
Right now, PFN_UP(trampoline_end - trampoline_start) is 2, causing more than
expected of the trampoline to be mapped. Cut it down just the single page it
ought to be.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
We have since gained a better location for it to live.
Fix up the includes while doing so. I don't recall why we had kernel.h but
it's definitely stale now.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
A CALL with 0 displacement is handled specially, and is why this logic
functions even with CET Shadow Stacks active. Nevertheless a RIP-relative LEA
is the more normal way of doing this in 64bit code.
The retrieval of flags modifies the stack pointer so needs to state a
dependency on the stack pointer. Despite it's name, ASM_CALL_CONSTRAINT is
the way to do this.
read_sreg() forces the answer through a register, causing code generation of
the form:
mov %gs, %eax
mov %eax, %eax
mov %rax, 0x140(%rsi)
Encode the reads directly with a memory operand. This results in a 16bit
store instead of an 64bit store, but the backing memory is zeroed.
While cleaning this up, drop one piece of trailing whitespace.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Use %pd, and state what the unknown is. As it's an enum, it's a signed type.
Also drop one piece of trailing whitespace.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
When PV is enabled, entry_int80 needs to be DPL3, not DPL0.
This, combined with a QEMU bug which incorrectly calculates the error
code (fix submitted separately), causes the XSA-259 PoC to fail with:
--- Xen Test Framework ---
Environment: PV 64bit (Long mode 4 levels)
XSA-259 PoC
Error: Unexpected fault 0x800d0802, #GP[IDT[256]]
Test result: ERROR
Fixes: 3da2149cf4 ("x86/IDT: Generate bsp_idt[] at build time")
Reported-by: Luca Fancellu <luca.fancellu@arm.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Tested-by: Luca Fancellu <luca.fancellu@arm.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
The use of "legitimately" mixes the concepts of "it was designed to do this"
and "it was correct to do this".
The latter in particular can go stale. "intended" is a better way of phrasing
this.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
When a watchdog fires, the domain is crashed and can't dump any state.
Xen allows a domain to have two separate watchdogs. Therefore, for a
domain running multiple watchdogs (e.g. one based around network, one
for disk), it is important for diagnostics to know which watchdog
fired.
As the printk() is in a timer callback, this is a bit awkward to
arrange, but there are 12 spare bits in the bottom of the domain
pointer owing to its alignment.
Reuse these bits to encode the watchdog id too, so the one which fired
is identified when the domain is crashed.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
As soon as the the domain is in the domlist, it can be queried via various
means, ahead of being fully constructed. Ensure it has the toolstack-given
UUID prior to becoming visible.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
We've already scanned features by the time init_e820() is called. Remove the
cpuid() calls.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
The exact behaviour of LVTERR interrupt generation is implementation
specific.
* Newer Intel CPUs generate an interrupt when pending_esr becomes
nonzero.
* Older Intel and all AMD CPUs generate an interrupt when any
individual bit in pending_esr becomes nonzero.
Neither vendor documents their behaviour very well. Xen implements
the per-bit behaviour and has done since support was added.
Importantly, the per-bit behaviour can be expressed using the atomic
operations available in the x86 architecture, whereas the
former (interrupt only on pending_esr becoming nonzero) cannot.
With vlapic->hw.pending_esr held outside of the main LAPIC register page,
it's much easier to use atomic operations.
Use xchg() in vlapic_reg_write(), and *set_bit() in vlapic_error().
The only interesting change is that vlapic_error() now needs to take a
single bit only, rather than a mask, but this fine for all current
callers and forseable changes.
No change from a guests perspective.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Xen currently presents APIC_ESR to guests as a simple read/write register.
This is incorrect. The SDM states:
The ESR is a write/read register. Before attempt to read from the ESR,
software should first write to it. (The value written does not affect the
values read subsequently; only zero may be written in x2APIC mode.) This
write clears any previously logged errors and updates the ESR with any
errors detected since the last write to the ESR.
Introduce a new pending_esr field in hvm_hw_lapic.
Update vlapic_error() to accumulate errors here, and extend vlapic_reg_write()
to discard the written value and transfer pending_esr into APIC_ESR. Reads
are still as before.
Importantly, this means that guests no longer destroys the ESR value it's
looking for in the LVTERR handler when following the SDM instructions.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
... to be more concise, and to match our other entrypoints into Xen.
In acpi_sleep_prepare(), calculate bootsym_phys() once, which GCC seems
unwilling to of it's own accord.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
There are two issues. First, pi_test_and_clear_on() pulls the cache-line to
the CPU and dirties it even if there's nothing outstanding, and second,
bitmap_for_each() is O(256) when O(8) would do, and would avoid multiple
atomic updates to the same IRR word.
Rewrite it from scratch, explaining what's going on at each step.
Bloat-o-meter reports 177 -> 145 (net -32), but real improvement is the
removal of calls to __find_{first,next}_bit() hidden behind bitmap_for_each().
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
struct domain is always a page aligned allocation. Update it's type to
reflect this, so we can safely reuse the lower bits in the pointer for
auxiliary information.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
With most of pv_trap_init() being done at build time, opening of NMI_SOFTIRQ
can be a regular initcall, simplifying trap_init().
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Now that bsp_idt[] is constructed at build time, we do not need to manually
initialise it in init_idt_traps() and trap_init().
When swapping the early pagefault handler for the normal one, switch to using
_update_gate_addr_lower() as we do on the kexec path for NMI and #MC.
This in turn allows us to drop set_{intr,swint}_gate() and the underlying
infrastructure. It also lets us drop autogen_entrypoints[] and that
underlying infrastructure.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
... rather than dynamically at boot time. Aside from less runtime overhead,
this approach is less fragile than the preexisting autogen stubs mechanism.
We can manage this with some linker calculations. See patch comments for full
details.
For simplicity, we create a new set of entry stubs here, and clean up the old
ones in the subsequent patch. bsp_idt[] needs to move from .bss to .data.
No functional change yet; the boot path still (re)writes bsp_idt[] at this
juncture.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
This can be a plain per_cpu() variable, and __read_mostly seeing as it's
allocated once and not touched again during it's lifetime.
This removes a NR_CPU's sized structure, and improves NUMA locality of access
for both the the VT-x and SVM context switch paths.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Having variables named idt_table[] and idt_tables[] is not ideal.
Use X86_IDT_VECTORS and remove IDT_ENTRIES. State the size of bsp_idt[] in
idt.h so that load_system_tables() and cpu_smpboot_alloc() can use sizeof()
rather than opencoding the calculation.
Move the variable into a new traps-setup.c, to make a start at splitting
traps.c in half.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
It has been discovered that VIRQ_ARGO is a 3rd type of VIRQ. Also, recent
work has prevented global VIRQs from being stolen from the owning domain.
Rewrite the description of VIRQ classifications. Drop the (DOM0) comment from
the global VIRQs; it's not been true for a long time.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
* Add cf_check annotation to cmp_patch_id() used by bsearch().
* Add U suffix to the K[] table to fix MISRA Rule 7.2 violations.
Fixes: 372af52441 ("xen/lib: Introduce SHA2-256")
Fixes: 630e8875ab ("x86/ucode: Perform extra SHA2 checks on AMD Fam17h/19h microcode")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
A future change will need to calculate SHA2-256 digests. Introduce an
implementation in lib/, derived from Trenchboot which itself is derived from
Linux.
In order to be useful to other architectures, it is careful with endianness
and misaligned accesses as well as being more MISRA friendly, but is only
wired up for x86 in the short term.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Observant readers may have noticed that the FRED spec has another 8 bits of
space reserved immediately following the vector field.
Make the existing constant more precise.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Logic concerning the IDT is somewhat different to the other system tables, and
in particular ought not to be in asm/processor.h. Collect it together a new
header.
While doing so, make a few minor adjustments:
* Make set_ist() use volatile rather than ACCESS_ONCE(), as
_write_gate_lower() already does, removing the need for xen/lib.h.
* Move the BUILD_BUG_ON() from subarch_percpu_traps_init() into mm.c's
build_assertions(), rather than including idt.h into x86_64/traps.c.
* Drop UL from IST constants.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
FRED support involves quite a lot of header file shuffling and cleanup. Start
by sorting the includes of impacted files, and dropping duplciates.
domain.c: Double asm/spec_ctrl.h
power.c: Double xen/sched.h
setup.c: Double xen/serial.h
mm.c: Double xen/mm.h
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
The final hunk is `(struct vcpu *)v` in disguise, expressed using a runtime
pointer chase through memory and a technicality of the C type system to work
around the fact that get_hvm_registers() strictly requires a mutable pointer.
For anyone interested, this is one reason why C cannot optimise any reads
across sequence points, even for a function purporting to take a const object.
Anyway, have the function correctly state that it needs a mutable vcpu. All
callers have a mutable vCPU to hand, and it removes the runtime pointer chase
in x86.
Make one style adjustment in ARM while adjusting the parameter type.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
"find upstream bridge" is surprisingly jarring in context, considering that's
the name of the function who's return value we're testing.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
The single user wants this the sane way around. Write it as a normal static
inline just like rspin_lock().
Fixes: cc3e8df542 ("xen/spinlock: add rspin_[un]lock_irq[save|restore]()")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
This ought to have been part of the original patch, so as to avoid breaking
CI.
Fixes: 31c0d6fdf4 ("xen/bsearch: Split out of lib.h into it's own header")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Nicola Vetrini <nicola.vetrini@bugseng.com>
There are currently two users, and lib.h is included everywhere.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Julien Grall <jgrall@amazon.com>
Also enable -fno-sanitize=alignment like x86 since support for unaligned
accesses is guaranteed by the ISA and the existing OPAL setup code
relies on it.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Shawn Anastasio <sanastasio@raptorengineering.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
They are out of place in traps.c, and only have a single caller each. Make
them static inside msr.c.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
It's out of place in traps.c, and only has a single caller. Make it static
inside cpuid.c.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
This was true in the K10 days, but even back then the match registers were
really payload data rather than header data.
But, it's really model specific data, and these days typically part of the
signature, so is random data for all intents and purposes.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
microcode_update{,_helper}() are overly generic names in a file that has
multiple update routines and helper functions contexts.
Rename microcode_update() to ucode_update_hcall() so it explicitly identifies
itself as hypercall context, and rename microcode_update_helper() to
ucode_update_hcall_cont() to make it clear it is in continuation context.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>
First, rename $TAG to $COMMITTISH. We already pass tags, branches (well, only
master) and full SHAs into this script.
Xen uses master for QEMU_UPSTREAM_REVISION, and has done for other trees too
in the path. Apparently we've never specified a different branch, because the
git-clone rune only pulls in the master branch; it does not pull in diverging
branches.
Fix this by performing an explicit fetch of the $COMMITTISH, then checking out
the dummy branch from the FETCH_HEAD.
Suggested-by: Jason Andryuk <jason.andryuk@amd.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Reviewed-by: Jason Andryuk <jason.andryuk@amd.com>
Now that Xen has a real vprintk(), there's no need to opencode it locally with
vsnprintf(). Redirect the debug routines to the real {v,}printk() and drop
the local acpi_os_{v,}printf() implementations.
Amongst other things, this removes one arbitrary limit on message size, as
well as removing a 512 byte static buffer that ought to have been in
__initdata given that is private to an __init function.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
For ABIs which pass parameters by register (all cases that we compile Xen
for), inserting new arguments on the left hand side involves shuffling all
other parameters along by one register whereas appending a new argument
doesn't involve shuffling of existing registers.
Reorder vprintk_common()'s prefix parameter to being last. This is a marginal
improvement on all architectures:
Function old new delta
vprintk 18 12 -6 x86
vprintk 32 24 -8 arm32
vprintk 52 48 -4 arm64
vprintk 52 48 -4 riscv64
vprintk 80 72 -8 ppc64
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
There is a corner case in the VMRUN instruction where its INTR_SHADOW state
leaks into guest state if a VMExit occurs before the VMRUN is complete. An
example of this could be taking #NPF due to event injection.
Xen can safely execute STI anywhere between CLGI and VMRUN, as CLGI blocks
external interrupts too. However, an exception (while fatal) will appear to
be in an irqs-on region (as GIF isn't considered), so position the STI after
the speculation actions but prior to the GPR pops.
Link: https://lore.kernel.org/all/CADH9ctBs1YPmE4aCfGPNBwA10cA8RuAk2gO7542DjMZgs4uzJQ@mail.gmail.com/
Fixes: 66b245d9ea ("SVM: limit GIF=0 region")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
The panic() function uses a static buffer to format its arguments into, simply
to emit the result via printk("%s", buf). This buffer is not large enough for
some existing users in Xen. e.g.:
(XEN) ****************************************
(XEN) Panic on CPU 0:
(XEN) Invalid device tree blob at physical address 0x46a00000.
(XEN) The DTB must be 8-byte aligned and must not exceed 2 MB in size.
(XEN)
(XEN) Plea****************************************
The remainder of this particular message is 'e check your bootloader.', but
has been inherited by RISC-V from ARM.
It is also pointless double buffering. Implement vprintk() beside printk(),
and use it directly rather than rendering into a local buffer, removing it as
one source of message limitation.
This marginally simplifies panic(), and drops a global used-once buffer.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
While fixing some common/arch boundaries for UBSAN support on other
architectures, the following debugging patch:
diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c
index c1f2d1b89d43..58d1d048d339 100644
--- a/xen/arch/arm/setup.c
+++ b/xen/arch/arm/setup.c
@@ -504,6 +504,8 @@ void asmlinkage __init start_xen(unsigned long fdt_paddr)
system_state = SYS_STATE_active;
+ dump_execution_state();
+
for_each_domain( d )
domain_unpause_by_systemcontroller(d);
failed with:
(XEN) *** Serial input to DOM0 (type 'CTRL-a' three times to switch input)
(XEN) CPU0: Unexpected Trap: Undefined Instruction
(XEN) ----[ Xen-4.20-rc arm32 debug=n Not tainted ]----
(XEN) CPU: 0
<snip>
(XEN)
(XEN) ****************************************
(XEN) Panic on CPU 0:
(XEN) CPU0: Unexpected Trap: Undefined Instruction
(XEN) ****************************************
This is because the condition for init text is wrong. While there's nothing
interesting from that point onwards in start_xen(), it's also wrong for
livepatches too.
Use is_active_kernel_text() which is the correct test for this purpose, and is
aware of init and livepatch regions as well as their lifetimes.
Fixes: 3e802c6ca1 ("xen/arm: Correctly support WARN_ON")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
RISC-V has less complicated headers, so update ubsan.c to pull in everything
it needs. Provide dump_execution_state(), and update the printk() message to
make it more obvious that it's an outstanding task.
As with commit 8ef2ac727e ("automation: enable UBSAN for debug tests"),
enable UBSAN in RISC-V testing too.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
JAL has a maximium displacement of 2M. To branch further, it needs pairing
with an AUIPC instruction. CALL is a pseudoinstruction which allows the
linker to pick the appropriate sequence when relaxations are enabled.
This avoids a build failure of the form:
prelink.o: in function `start':
xen/xen/arch/riscv/riscv64/head.S:28:(.text.header+0x2c):
relocation truncated to fit: R_RISCV_JAL against symbol `calc_phys_offset' defined in .init.text section in prelink.o
make[3]: *** [arch/riscv/Makefile:18: xen-syms] Error 1
when Xen gets large enough, e.g. with CONFIG_UBSAN enabled.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>
Logic using performance counters needs to look at
MSR_MISC_ENABLE.PERF_AVAILABLE before touching any other resources.
When virtualised under ESX, Xen dies with a #GP fault trying to read
MSR_CORE_PERF_GLOBAL_CTRL.
Factor this logic out into a separate function (it's already too squashed to
the RHS), and insert a check of MSR_MISC_ENABLE.PERF_AVAILABLE.
This also avoids setting X86_FEATURE_ARCH_PERFMON if MSR_MISC_ENABLE says that
PERF is unavailable, although oprofile (the only consumer of this flag)
cross-checks too.
Fixes: 6bdb965178 ("x86/intel: ensure Global Performance Counter Control is setup correctly")
Reported-by: Jonathan Katz <jonathan.katz@aptar.com>
Link: https://xcp-ng.org/forum/topic/10286/nesting-xcp-ng-on-esx-8
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
Tested-by: Jonathan Katz <jonathan.katz@aptar.com>
public/arch-ppc.h contains two adjacent #ifndef __ASSEMBLY__ blocks.
With these merged, it becomes very obvious that there's a duplicate
definition of xen_ulong_t, which is also noticed by the docs build:
/usr/bin/perl -w /local/xen.git/docs/xen-headers -O html/hypercall/ppc \
-T 'arch-ppc - Xen public headers' \
-X arch-arm -X arch-riscv -X arch-x86_32 -X arch-x86_64 \
-X xen-arm -X xen-riscv -X xen-x86_32 -X xen-x86_64 \
-X arch-x86 \
/local/xen.git/docs/../xen include/public include/xen/errno.h
include/public/memory.h:63: multiple definitions of Typedef xen_ulong_t: include/public/arch-ppc.h:55
include/public/memory.h:63: multiple definitions of Typedef xen_ulong_t: include/public/arch-ppc.h:61
include/public/memory.h:63: multiple definitions of Typedef xen_ulong_t: include/public/arch-ppc.h:61
include/public/memory.h:63: multiple definitions of Typedef xen_ulong_t: include/public/arch-ppc.h:55
Drop the second typedef. Finally, annotate the #endif so it's clear
what it refers to.
Fixes: 08c192cc11 ("xen/ppc: Add public/arch-ppc.h")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Shawn Anastasio <sanastasio@raptorengineering.com>
* Strip trailing whitspace.
* Remove PRIperfc. It has never been used and isn't useful.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
This is mostly for the removal of xen/lib.h and xen/smp.h from perfc.h. All
that is needed is xen/macros.h.
Trim and sort the includes for perfc.c too. There's no need for smp.h,
keyhandler.h or mm.h, but cpumask.h is needed.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
... and hook it up for RISC-V and PPC.
On RISC-V at least, no combination of headers pulls in errno.h, so include it
explicitly.
Guard the hypercalls array declaration based on NR_hypercalls existing. This
is sufficient to get PERF_COUNTERS fully working on RISC-V and PPC, so drop
the randconfig override.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Oleksii Kurohcko <oleksii.kurochko@gmail.com>
These were only ever used by the IA64 port, which was droped in commit
570c311ca2 ("remove ia64") in 2012.
Remove them, and clean up the arm/x86 stub headers.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
Clang, of at least verion 17 complains:
arch/x86/pv/hypercall.c:30:10: error: variable 'eax' is used uninitialized
whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
30 | if ( !compat )
| ^~~~~~~
arch/x86/pv/hypercall.c:87:29: note: uninitialized use occurs here
87 | perfc_incra(hypercalls, eax);
| ^~~
This function is forced always_inline to cause compat to be
constant-propagated through, but that is only a heuristic to try and get the
compiler to do what we want, not a gurantee that it does.
Clang doesn't appear to be able to see that the only case where compat is
true (and therefore the if() is false) is when there's an else clause on the
end which sets eax too.
Initialise eax to -1, which ought to be optimised out, but if for whatever
reason it happens not to be, then perfc_incra() will fail it's bounds check
and do nothing.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
AMD have always used the architectural MSRs for LER. As the first processor
to support LER was the K7 (which was 32bit), we can assume it's presence
unconditionally in 64bit mode.
Intel are about to run out of space in Family 6 and start using 19. It is
only the Pentium 4 which uses non-architectural LER MSRs.
percpu_traps_init(), which runs on every CPU, contains a lot of code which
should be init-only, and is the only reason why opt_ler can't be in initdata.
Write a brand new init_ler() which expects all future Intel and AMD CPUs to
continue using the architectural MSRs, and does all setup together. Call it
from trap_init(), and remove the setup logic percpu_traps_init() except for
the single path configuring MSR_IA32_DEBUGCTLMSR.
Leave behind a warning if the user asked for LER and Xen couldn't enable it.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
AMD have updated the SRSO whitepaper[1] with further information. These
features exist on AMD Zen5 CPUs and are necessary for Xen to use.
The two features are in principle unrelated:
* SRSO_U/S_NO is an enumeration saying that SRSO attacks can't cross the
User(CPL3) / Supervisor(CPL<3) boundary. i.e. Xen don't need to use
IBPB-on-entry for PV64. PV32 guests are explicitly unsupported for
speculative issues, and excluded from consideration for simplicity.
* SRSO_MSR_FIX is an enumeration identifying that the BP_SPEC_REDUCE bit is
available in MSR_BP_CFG. When set, SRSO attacks can't cross the host/guest
boundary. i.e. Xen don't need to use IBPB-on-entry for HVM.
Extend ibpb_calculations() to account for these when calculating
opt_ibpb_entry_{pv,hvm} defaults. Add a `bp-spec-reduce=<bool>` option to
control the use of BP_SPEC_REDUCE, with it active by default.
Because MSR_BP_CFG is core-scoped with a race condition updating it, repurpose
amd_check_erratum_1485() into amd_check_bp_cfg() and calculate all updates at
once.
Xen also needs to to advertise SRSO_U/S_NO to guests to allow the guest kernel
to skip SRSO mitigations too:
* This is trivial for HVM guests. It is also is accurate for PV32 guests
too, but we have already excluded them from consideration, and do so again
here to simplify the policy logic.
* As written, SRSO_U/S_NO does not help for the PV64 user->kernel boundary.
However, after discussing with AMD, an implementation detail of having
BP_SPEC_REDUCE active causes the PV64 user->kernel boundary to have the
property described by SRSO_U/S_NO, so we can advertise SRSO_U/S_NO to
guests when the BP_SPEC_REDUCE precondition is met.
Finally, fix a typo in the SRSO_NO's comment.
[1] https://www.amd.com/content/dam/amd/en/documents/corporate/cr/speculative-return-stack-overflow-whitepaper.pdf
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Roger Pau Monné <roger.pau@citrix.com>
... which converts binary content to hex automatically.
Update libxl to match. No API/ABI change.
This removes a latent libxl bug for cases when the buildid is longer than 4092
bytes.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
... which uses XENVER_extraversion2.
In order to do this sensibly, use manual hypercall buffer handling. Not only
does this avoid an extra bounce buffer (we need to strip the xen_varbuf_t
header anyway), it's also shorter and easlier to follow.
Update libxl and the ocaml stubs to match. No API/ABI change in either.
With this change made, `xl info` can now correctly access a >15 char
extraversion:
# xl info xen_version
4.18-unstable+REALLY LONG EXTRAVERSION
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
kexec-tools uses xc_version(), meaning that it is not a private API. As we're
going to extend the functionality substantially, move it to its own file.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
struct xen_build_id and struct xen_varbuf are identical from an ABI point of
view, so XENVER_build_id can reuse xenver_varbuf_op() rather than having it's
own almost identical copy of the logic.
No functional change.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
In XenServer, we have encountered problems caused by both XENVER_extraversion
and XENVER_commandline having fixed bounds.
More than just the invariant size, the APIs/ABIs also broken by typedef-ing an
array, and using an unqualified 'char' which has implementation-specific
signed-ness.
Provide brand new ops, which are capable of expressing variable length
strings, and mark the older ops as broken.
This fixes all issues around XENVER_extraversion being longer than 15 chars.
Further work beyond just this API is needed to remove other assumptions about
XENVER_commandline being 1023 chars long.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jason Andryuk <jandryuk@gmail.com>
---
Non-technical objections to this patch were raised, and subsequently rejected
by a community wide vote. The results of the vote have not been shared with
the community at the time of committing.
Andrew Cooper