| Age | Commit message (Collapse) | Author |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Thomas Gleixner:
- Move the command line preparation and the early command line parsing
earlier so that the command line parameters which affect
early_reserve_memory(), e.g. efi=nosftreserve, are taken into
account. This was broken when the invocation of
early_reserve_memory() was moved recently.
- Use an atomic type for the SGX page accounting, which is read and
written locklessly, to plug various race conditions related to it.
* tag 'x86-urgent-2021-11-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/sgx: Fix free page accounting
x86/boot: Pull up cmdline preparation and early param parsing
|
|
The SGX driver maintains a single global free page counter,
sgx_nr_free_pages, that reflects the number of free pages available
across all NUMA nodes. Correspondingly, a list of free pages is
associated with each NUMA node and sgx_nr_free_pages is updated
every time a page is added or removed from any of the free page
lists. The main usage of sgx_nr_free_pages is by the reclaimer
that runs when it (sgx_nr_free_pages) goes below a watermark
to ensure that there are always some free pages available to, for
example, support efficient page faults.
With sgx_nr_free_pages accessed and modified from a few places
it is essential to ensure that these accesses are done safely but
this is not the case. sgx_nr_free_pages is read without any
protection and updated with inconsistent protection by any one
of the spin locks associated with the individual NUMA nodes.
For example:
CPU_A CPU_B
----- -----
spin_lock(&nodeA->lock); spin_lock(&nodeB->lock);
... ...
sgx_nr_free_pages--; /* NOT SAFE */ sgx_nr_free_pages--;
spin_unlock(&nodeA->lock); spin_unlock(&nodeB->lock);
Since sgx_nr_free_pages may be protected by different spin locks
while being modified from different CPUs, the following scenario
is possible:
CPU_A CPU_B
----- -----
{sgx_nr_free_pages = 100}
spin_lock(&nodeA->lock); spin_lock(&nodeB->lock);
sgx_nr_free_pages--; sgx_nr_free_pages--;
/* LOAD sgx_nr_free_pages = 100 */ /* LOAD sgx_nr_free_pages = 100 */
/* sgx_nr_free_pages-- */ /* sgx_nr_free_pages-- */
/* STORE sgx_nr_free_pages = 99 */ /* STORE sgx_nr_free_pages = 99 */
spin_unlock(&nodeA->lock); spin_unlock(&nodeB->lock);
In the above scenario, sgx_nr_free_pages is decremented from two CPUs
but instead of sgx_nr_free_pages ending with a value that is two less
than it started with, it was only decremented by one while the number
of free pages were actually reduced by two. The consequence of
sgx_nr_free_pages not being protected is that its value may not
accurately reflect the actual number of free pages on the system,
impacting the availability of free pages in support of many flows.
The problematic scenario is when the reclaimer does not run because it
believes there to be sufficient free pages while any attempt to allocate
a page fails because there are no free pages available. In the SGX driver
the reclaimer's watermark is only 32 pages so after encountering the
above example scenario 32 times a user space hang is possible when there
are no more free pages because of repeated page faults caused by no
free pages made available.
The following flow was encountered:
asm_exc_page_fault
...
sgx_vma_fault()
sgx_encl_load_page()
sgx_encl_eldu() // Encrypted page needs to be loaded from backing
// storage into newly allocated SGX memory page
sgx_alloc_epc_page() // Allocate a page of SGX memory
__sgx_alloc_epc_page() // Fails, no free SGX memory
...
if (sgx_should_reclaim(SGX_NR_LOW_PAGES)) // Wake reclaimer
wake_up(&ksgxd_waitq);
return -EBUSY; // Return -EBUSY giving reclaimer time to run
return -EBUSY;
return -EBUSY;
return VM_FAULT_NOPAGE;
The reclaimer is triggered in above flow with the following code:
static bool sgx_should_reclaim(unsigned long watermark)
{
return sgx_nr_free_pages < watermark &&
!list_empty(&sgx_active_page_list);
}
In the problematic scenario there were no free pages available yet the
value of sgx_nr_free_pages was above the watermark. The allocation of
SGX memory thus always failed because of a lack of free pages while no
free pages were made available because the reclaimer is never started
because of sgx_nr_free_pages' incorrect value. The consequence was that
user space kept encountering VM_FAULT_NOPAGE that caused the same
address to be accessed repeatedly with the same result.
Change the global free page counter to an atomic type that
ensures simultaneous updates are done safely. While doing so, move
the updating of the variable outside of the spin lock critical
section to which it does not belong.
Cc: stable@vger.kernel.org
Fixes: 901ddbb9ecf5 ("x86/sgx: Add a basic NUMA allocation scheme to sgx_alloc_epc_page()")
Suggested-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Jarkko Sakkinen <jarkko@kernel.org>
Link: https://lkml.kernel.org/r/a95a40743bbd3f795b465f30922dde7f1ea9e0eb.1637004094.git.reinette.chatre@intel.com
|
|
Explicitly check for MSR_HYPERCALL and MSR_VP_INDEX support when probing
for running as a Hyper-V guest instead of waiting until hyperv_init() to
detect the bogus configuration. Add messages to give the admin a heads
up that they are likely running on a broken virtual machine setup.
At best, silently disabling Hyper-V is confusing and difficult to debug,
e.g. the kernel _says_ it's using all these fancy Hyper-V features, but
always falls back to the native versions. At worst, the half baked setup
will crash/hang the kernel.
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20211104182239.1302956-3-seanjc@google.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
- Add the model number of a new, Raptor Lake CPU, to intel-family.h
- Do not log spurious corrected MCEs on SKL too, due to an erratum
- Clarify the path of paravirt ops patches upstream
- Add an optimization to avoid writing out AMX components to sigframes
when former are in init state
* tag 'x86_urgent_for_v5.16_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/cpu: Add Raptor Lake to Intel family
x86/mce: Add errata workaround for Skylake SKX37
MAINTAINERS: Add some information to PARAVIRT_OPS entry
x86/fpu: Optimize out sigframe xfeatures when in init state
|
|
Errata SKX37 is word-for-word identical to the other errata listed in
this workaround. I happened to notice this after investigating a CMCI
storm on a Skylake host. While I can't confirm this was the root cause,
spurious corrected errors does sound like a likely suspect.
Fixes: 2976908e4198 ("x86/mce: Do not log spurious corrected mce errors")
Signed-off-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20211029205759.GA7385@codemonkey.org.uk
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core updates from Greg KH:
"Here is the big set of driver core changes for 5.16-rc1.
All of these have been in linux-next for a while now with no reported
problems.
Included in here are:
- big update and cleanup of the sysfs abi documentation files and
scripts from Mauro. We are almost at the place where we can
properly check that the running kernel's sysfs abi is documented
fully.
- firmware loader updates
- dyndbg updates
- kernfs cleanups and fixes from Christoph
- device property updates
- component fix
- other minor driver core cleanups and fixes"
* tag 'driver-core-5.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (122 commits)
device property: Drop redundant NULL checks
x86/build: Tuck away built-in firmware under FW_LOADER
vmlinux.lds.h: wrap built-in firmware support under FW_LOADER
firmware_loader: move struct builtin_fw to the only place used
x86/microcode: Use the firmware_loader built-in API
firmware_loader: remove old DECLARE_BUILTIN_FIRMWARE()
firmware_loader: formalize built-in firmware API
component: do not leave master devres group open after bind
dyndbg: refine verbosity 1-4 summary-detail
gpiolib: acpi: Replace custom code with device_match_acpi_handle()
i2c: acpi: Replace custom function with device_match_acpi_handle()
driver core: Provide device_match_acpi_handle() helper
dyndbg: fix spurious vNpr_info change
dyndbg: no vpr-info on empty queries
dyndbg: vpr-info on remove-module complete, not starting
device property: Add missed header in fwnode.h
Documentation: dyndbg: Improve cli param examples
dyndbg: Remove support for ddebug_query param
dyndbg: make dyndbg a known cli param
dyndbg: show module in vpr-info in dd-exec-queries
...
|
|
tl;dr: AMX state is ~8k. Signal frames can have space for this
~8k and each signal entry writes out all 8k even if it is zeros.
Skip writing zeros for AMX to speed up signal delivery by about
4% overall when AMX is in its init state.
This is a user-visible change to the sigframe ABI.
== Hardware XSAVE Background ==
XSAVE state components may be tracked by the processor as being
in their initial configuration. Software can detect which
features are in this configuration by looking at the XSTATE_BV
field in an XSAVE buffer or with the XGETBV(1) instruction.
Both the XSAVE and XSAVEOPT instructions enumerate features s
being in the initial configuration via the XSTATE_BV field in the
XSAVE header, However, XSAVEOPT declines to actually write
features in their initial configuration to the buffer. XSAVE
writes the feature unconditionally, regardless of whether it is
in the initial configuration or not.
Basically, XSAVE users never need to inspect XSTATE_BV to
determine if the feature has been written to the buffer.
XSAVEOPT users *do* need to inspect XSTATE_BV. They might also
need to clear out the buffer if they want to make an isolated
change to the state, like modifying one register.
== Software Signal / XSAVE Background ==
Signal frames have historically been written with XSAVE itself.
Each state is written in its entirety, regardless of being in its
initial configuration.
In other words, the signal frame ABI uses the XSAVE behavior, not
the XSAVEOPT behavior.
== Problem ==
This means that any application which has acquired permission to
use AMX via ARCH_REQ_XCOMP_PERM will write 8k of state to the
signal frame. This 8k write will occur even when AMX was in its
initial configuration and software *knows* this because of
XSTATE_BV.
This problem also exists to a lesser degree with AVX-512 and its
2k of state. However, AVX-512 use does not require
ARCH_REQ_XCOMP_PERM and is more likely to have existing users
which would be impacted by any change in behavior.
== Solution ==
Stop writing out AMX xfeatures which are in their initial state
to the signal frame. This effectively makes the signal frame
XSAVE buffer look as if it were written with a combination of
XSAVEOPT and XSAVE behavior. Userspace which handles XSAVEOPT-
style buffers should be able to handle this naturally.
For now, include only the AMX xfeatures: XTILE and XTILEDATA in
this new behavior. These require new ABI to use anyway, which
makes their users very unlikely to be broken. This XSAVEOPT-like
behavior should be expected for all future dynamic xfeatures. It
may also be extended to legacy features like AVX-512 in the
future.
Only attempt this optimization on systems with dynamic features.
Disable dynamic feature support (XFD) if XGETBV1 is unavailable
by adding a CPUID dependency.
This has been measured to reduce the *overall* cycle cost of
signal delivery by about 4%.
Fixes: 2308ee57d93d ("x86/fpu/amx: Enable the AMX feature in 64-bit mode")
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: "Chang S. Bae" <chang.seok.bae@intel.com>
Link: https://lore.kernel.org/r/20211102224750.FA412E26@davehans-spike.ostc.intel.com
|
|
Pull drm updates from Dave Airlie:
"Summary below. i915 starts to add support for DG2 GPUs, enables DG1
and ADL-S support by default, lots of work to enable DisplayPort 2.0
across drivers. Lots of documentation updates and fixes across the
board.
core:
- improve dma_fence, lease and resv documentation
- shmem-helpers: allocate WC pages on x86, use vmf_insert_pin
- sched fixes/improvements
- allow empty drm leases
- add dma resv iterator
- add more DP 2.0 headers
- DP MST helper improvements for DP2.0
dma-buf:
- avoid warnings, remove fence trace macros
bridge:
- new helper to get rid of panels
- probe improvements for it66121
- enable DSI EOTP for anx7625
fbdev:
- efifb: release runtime PM on destroy
ttm:
- kerneldoc switch
- helper to clear all DMA mappings
- pool shrinker optimizaton
- remove ttm_tt_destroy_common
- update ttm_move_memcpy for async use
panel:
- add new panel-edp driver
amdgpu:
- Initial DP 2.0 support
- Initial USB4 DP tunnelling support
- Aldebaran MCE support
- Modifier support for DCC image stores for GFX 10.3
- Display rework for better FP code handling
- Yellow Carp/Cyan Skillfish updates
- Cyan Skillfish display support
- convert vega/navi to IP discovery asic enumeration
- validate IP discovery table
- RAS improvements
- Lots of fixes
i915:
- DG1 PCI IDs + LMEM discovery/placement
- DG1 GuC submission by default
- ADL-S PCI IDs updated + enabled by default
- ADL-P (XE_LPD) fixed and updates
- DG2 display fixes
- PXP protected object support for Gen12 integrated
- expose multi-LRC submission interface for GuC
- export logical engine instance to user
- Disable engine bonding on Gen12+
- PSR cleanup
- PSR2 selective fetch by default
- DP 2.0 prep work
- VESA vendor block + MSO use of it
- FBC refactor
- try again to fix fast-narrow vs slow-wide eDP training
- use THP when IOMMU enabled
- LMEM backup/restore for suspend/resume
- locking simplification
- GuC major reworking
- async flip VT-D workaround changes
- DP link training improvements
- misc display refactorings
bochs:
- new PCI ID
rcar-du:
- Non-contiguious buffer import support for rcar-du
- r8a779a0 support prep
omapdrm:
- COMPILE_TEST fixes
sti:
- COMPILE_TEST fixes
msm:
- fence ordering improvements
- eDP support in DP sub-driver
- dpu irq handling cleanup
- CRC support for making igt happy
- NO_CONNECTOR bridge support
- dsi: 14nm phy support for msm8953
- mdp5: msm8x53, sdm450, sdm632 support
stm:
- layer alpha + zpo support
v3d:
- fix Vulkan CTS failure
- support multiple sync objects
gud:
- add R8/RGB332/RGB888 pixel formats
vc4:
- convert to new bridge helpers
vgem:
- use shmem helpers
virtio:
- support mapping exported vram
zte:
- remove obsolete driver
rockchip:
- use bridge attach no connector for LVDS/RGB"
* tag 'drm-next-2021-11-03' of git://anongit.freedesktop.org/drm/drm: (1259 commits)
drm/amdgpu/gmc6: fix DMA mask from 44 to 40 bits
drm/amd/display: MST support for DPIA
drm/amdgpu: Fix even more out of bound writes from debugfs
drm/amdgpu/discovery: add SDMA IP instance info for soc15 parts
drm/amdgpu/discovery: add UVD/VCN IP instance info for soc15 parts
drm/amdgpu/UAPI: rearrange header to better align related items
drm/amd/display: Enable dpia in dmub only for DCN31 B0
drm/amd/display: Fix USB4 hot plug crash issue
drm/amd/display: Fix deadlock when falling back to v2 from v3
drm/amd/display: Fallback to clocks which meet requested voltage on DCN31
drm/amd/display: move FPU associated DCN301 code to DML folder
drm/amd/display: fix link training regression for 1 or 2 lane
drm/amd/display: add two lane settings training options
drm/amd/display: decouple hw_lane_settings from dpcd_lane_settings
drm/amd/display: implement decide lane settings
drm/amd/display: adopt DP2.0 LT SCR revision 8
drm/amd/display: FEC configuration for dpia links in MST mode
drm/amd/display: FEC configuration for dpia links
drm/amd/display: Add workaround flag for EDID read on certain docks
drm/amd/display: Set phy_mux_sel bit in dmub scratch register
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux
Pull hyperv updates from Wei Liu:
- Initial patch set for Hyper-V isolation VM support (Tianyu Lan)
- Fix a warning on preemption (Vitaly Kuznetsov)
- A bunch of misc cleanup patches
* tag 'hyperv-next-signed-20211102' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
x86/hyperv: Protect set_hv_tscchange_cb() against getting preempted
Drivers: hv : vmbus: Adding NULL pointer check
x86/hyperv: Remove duplicate include
x86/hyperv: Remove duplicated include in hv_init
Drivers: hv: vmbus: Remove unused code to check for subchannels
Drivers: hv: vmbus: Initialize VMbus ring buffer for Isolation VM
Drivers: hv: vmbus: Add SNP support for VMbus channel initiate message
x86/hyperv: Add ghcb hvcall support for SNP VM
x86/hyperv: Add Write/Read MSR registers via ghcb page
Drivers: hv: vmbus: Mark vmbus ring buffer visible to host in Isolation VM
x86/hyperv: Add new hvcall guest address host visibility support
x86/hyperv: Initialize shared memory boundary in the Isolation VM.
x86/hyperv: Initialize GHCB page in Isolation VM
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull seccomp updates from Kees Cook:
"These are x86-specific, but I carried these since they're also
seccomp-specific.
This flips the defaults for spec_store_bypass_disable and
spectre_v2_user from "seccomp" to "prctl", as enough time has passed
to allow system owners to have updated the defensive stances of their
various workloads, and it's long overdue to unpessimize seccomp
threads.
Extensive rationale and details are in Andrea's main patch.
Summary:
- set spec_store_bypass_disable & spectre_v2_user to prctl (Andrea Arcangeli)"
* tag 'seccomp-v5.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
x86: deduplicate the spectre_v2_user documentation
x86: change default to spec_store_bypass_disable=prctl spectre_v2_user=prctl
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 SGX updates from Borislav Petkov:
"Add a SGX_IOC_VEPC_REMOVE ioctl to the /dev/sgx_vepc virt interface
with which EPC pages can be put back into their uninitialized state
without having to reopen /dev/sgx_vepc, which could not be possible
anymore after startup due to security policies"
* tag 'x86_sgx_for_v5.16_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/sgx/virt: implement SGX_IOC_VEPC_REMOVE ioctl
x86/sgx/virt: extract sgx_vepc_remove_page
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 cpu updates from Borislav Petkov:
- Start checking a CPUID bit on AMD Zen3 which states that the CPU
clears the segment base when a null selector is written. Do the
explicit detection on older CPUs, zen2 and hygon specifically, which
have the functionality but do not advertize the CPUID bit. Factor in
the presence of a hypervisor underneath the kernel and avoid doing
the explicit check there which the HV might've decided to not
advertize for migration safety reasons, or similar.
- Add support for a new X86 CPU vendor: VORTEX. Needed for whitelisting
those CPUs in the hardware vulnerabilities detection
- Force the compiler to use rIP-relative addressing in the fallback
path of static_cpu_has(), in order to avoid unnecessary register
pressure
* tag 'x86_cpu_for_v5.16_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/cpu: Fix migration safety with X86_BUG_NULL_SEL
x86/CPU: Add support for Vortex CPUs
x86/umip: Downgrade warning messages to debug loglevel
x86/asm: Avoid adding register pressure for the init case in static_cpu_has()
x86/asm: Add _ASM_RIP() macro for x86-64 (%rip) suffix
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull RAS updates from Borislav Petkov:
- Get rid of a bunch of function pointers used in MCA land in favor of
normal functions. This is in preparation of making the MCA code
noinstr-aware
- When the kernel copies data from user addresses and it encounters a
machine check, a SIGBUS is sent to that process. Change this action
to either an -EFAULT which is returned to the user or a short write,
making the recovery action a lot more user-friendly
* tag 'ras_core_for_v5.16_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mce: Sort mca_config members to get rid of unnecessary padding
x86/mce: Get rid of the ->quirk_no_way_out() indirect call
x86/mce: Get rid of msr_ops
x86/mce: Get rid of machine_check_vector
x86/mce: Get rid of the mce_severity function pointer
x86/mce: Drop copyin special case for #MC
x86/mce: Change to not send SIGBUS error during copy from user
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fpu updates from Thomas Gleixner:
- Cleanup of extable fixup handling to be more robust, which in turn
allows to make the FPU exception fixups more robust as well.
- Change the return code for signal frame related failures from
explicit error codes to a boolean fail/success as that's all what the
calling code evaluates.
- A large refactoring of the FPU code to prepare for adding AMX
support:
- Distangle the public header maze and remove especially the
misnomed kitchen sink internal.h which is despite it's name
included all over the place.
- Add a proper abstraction for the register buffer storage (struct
fpstate) which allows to dynamically size the buffer at runtime
by flipping the pointer to the buffer container from the default
container which is embedded in task_struct::tread::fpu to a
dynamically allocated container with a larger register buffer.
- Convert the code over to the new fpstate mechanism.
- Consolidate the KVM FPU handling by moving the FPU related code
into the FPU core which removes the number of exports and avoids
adding even more export when AMX has to be supported in KVM.
This also removes duplicated code which was of course
unnecessary different and incomplete in the KVM copy.
- Simplify the KVM FPU buffer handling by utilizing the new
fpstate container and just switching the buffer pointer from the
user space buffer to the KVM guest buffer when entering
vcpu_run() and flipping it back when leaving the function. This
cuts the memory requirements of a vCPU for FPU buffers in half
and avoids pointless memory copy operations.
This also solves the so far unresolved problem of adding AMX
support because the current FPU buffer handling of KVM inflicted
a circular dependency between adding AMX support to the core and
to KVM. With the new scheme of switching fpstate AMX support can
be added to the core code without affecting KVM.
- Replace various variables with proper data structures so the
extra information required for adding dynamically enabled FPU
features (AMX) can be added in one place
- Add AMX (Advanced Matrix eXtensions) support (finally):
AMX is a large XSTATE component which is going to be available with
Saphire Rapids XEON CPUs. The feature comes with an extra MSR
(MSR_XFD) which allows to trap the (first) use of an AMX related
instruction, which has two benefits:
1) It allows the kernel to control access to the feature
2) It allows the kernel to dynamically allocate the large register
state buffer instead of burdening every task with the the extra
8K or larger state storage.
It would have been great to gain this kind of control already with
AVX512.
The support comes with the following infrastructure components:
1) arch_prctl() to
- read the supported features (equivalent to XGETBV(0))
- read the permitted features for a task
- request permission for a dynamically enabled feature
Permission is granted per process, inherited on fork() and
cleared on exec(). The permission policy of the kernel is
restricted to sigaltstack size validation, but the syscall
obviously allows further restrictions via seccomp etc.
2) A stronger sigaltstack size validation for sys_sigaltstack(2)
which takes granted permissions and the potentially resulting
larger signal frame into account. This mechanism can also be used
to enforce factual sigaltstack validation independent of dynamic
features to help with finding potential victims of the 2K
sigaltstack size constant which is broken since AVX512 support
was added.
3) Exception handling for #NM traps to catch first use of a extended
feature via a new cause MSR. If the exception was caused by the
use of such a feature, the handler checks permission for that
feature. If permission has not been granted, the handler sends a
SIGILL like the #UD handler would do if the feature would have
been disabled in XCR0. If permission has been granted, then a new
fpstate which fits the larger buffer requirement is allocated.
In the unlikely case that this allocation fails, the handler
sends SIGSEGV to the task. That's not elegant, but unavoidable as
the other discussed options of preallocation or full per task
permissions come with their own set of horrors for kernel and/or
userspace. So this is the lesser of the evils and SIGSEGV caused
by unexpected memory allocation failures is not a fundamentally
new concept either.
When allocation succeeds, the fpstate properties are filled in to
reflect the extended feature set and the resulting sizes, the
fpu::fpstate pointer is updated accordingly and the trap is
disarmed for this task permanently.
4) Enumeration and size calculations
5) Trap switching via MSR_XFD
The XFD (eXtended Feature Disable) MSR is context switched with
the same life time rules as the FPU register state itself. The
mechanism is keyed off with a static key which is default
disabled so !AMX equipped CPUs have zero overhead. On AMX enabled
CPUs the overhead is limited by comparing the tasks XFD value
with a per CPU shadow variable to avoid redundant MSR writes. In
case of switching from a AMX using task to a non AMX using task
or vice versa, the extra MSR write is obviously inevitable.
All other places which need to be aware of the variable feature
sets and resulting variable sizes are not affected at all because
they retrieve the information (feature set, sizes) unconditonally
from the fpstate properties.
6) Enable the new AMX states
Note, this is relatively new code despite the fact that AMX support
is in the works for more than a year now.
The big refactoring of the FPU code, which allowed to do a proper
integration has been started exactly 3 weeks ago. Refactoring of the
existing FPU code and of the original AMX patches took a week and has
been subject to extensive review and testing. The only fallout which
has not been caught in review and testing right away was restricted
to AMX enabled systems, which is completely irrelevant for anyone
outside Intel and their early access program. There might be dragons
lurking as usual, but so far the fine grained refactoring has held up
and eventual yet undetected fallout is bisectable and should be
easily addressable before the 5.16 release. Famous last words...
Many thanks to Chang Bae and Dave Hansen for working hard on this and
also to the various test teams at Intel who reserved extra capacity
to follow the rapid development of this closely which provides the
confidence level required to offer this rather large update for
inclusion into 5.16-rc1
* tag 'x86-fpu-2021-11-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (110 commits)
Documentation/x86: Add documentation for using dynamic XSTATE features
x86/fpu: Include vmalloc.h for vzalloc()
selftests/x86/amx: Add context switch test
selftests/x86/amx: Add test cases for AMX state management
x86/fpu/amx: Enable the AMX feature in 64-bit mode
x86/fpu: Add XFD handling for dynamic states
x86/fpu: Calculate the default sizes independently
x86/fpu/amx: Define AMX state components and have it used for boot-time checks
x86/fpu/xstate: Prepare XSAVE feature table for gaps in state component numbers
x86/fpu/xstate: Add fpstate_realloc()/free()
x86/fpu/xstate: Add XFD #NM handler
x86/fpu: Update XFD state where required
x86/fpu: Add sanity checks for XFD
x86/fpu: Add XFD state to fpstate
x86/msr-index: Add MSRs for XFD
x86/cpufeatures: Add eXtended Feature Disabling (XFD) feature bit
x86/fpu: Reset permission and fpstate on exec()
x86/fpu: Prepare fpu_clone() for dynamically enabled features
x86/fpu/signal: Prepare for variable sigframe length
x86/signal: Use fpu::__state_user_size for sigalt stack validation
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler updates from Thomas Gleixner:
- Revert the printk format based wchan() symbol resolution as it can
leak the raw value in case that the symbol is not resolvable.
- Make wchan() more robust and work with all kind of unwinders by
enforcing that the task stays blocked while unwinding is in progress.
- Prevent sched_fork() from accessing an invalid sched_task_group
- Improve asymmetric packing logic
- Extend scheduler statistics to RT and DL scheduling classes and add
statistics for bandwith burst to the SCHED_FAIR class.
- Properly account SCHED_IDLE entities
- Prevent a potential deadlock when initial priority is assigned to a
newly created kthread. A recent change to plug a race between cpuset
and __sched_setscheduler() introduced a new lock dependency which is
now triggered. Break the lock dependency chain by moving the priority
assignment to the thread function.
- Fix the idle time reporting in /proc/uptime for NOHZ enabled systems.
- Improve idle balancing in general and especially for NOHZ enabled
systems.
- Provide proper interfaces for live patching so it does not have to
fiddle with scheduler internals.
- Add cluster aware scheduling support.
- A small set of tweaks for RT (irqwork, wait_task_inactive(), various
scheduler options and delaying mmdrop)
- The usual small tweaks and improvements all over the place
* tag 'sched-core-2021-11-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (69 commits)
sched/fair: Cleanup newidle_balance
sched/fair: Remove sysctl_sched_migration_cost condition
sched/fair: Wait before decaying max_newidle_lb_cost
sched/fair: Skip update_blocked_averages if we are defering load balance
sched/fair: Account update_blocked_averages in newidle_balance cost
x86: Fix __get_wchan() for !STACKTRACE
sched,x86: Fix L2 cache mask
sched/core: Remove rq_relock()
sched: Improve wake_up_all_idle_cpus() take #2
irq_work: Also rcuwait for !IRQ_WORK_HARD_IRQ on PREEMPT_RT
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT
irq_work: Allow irq_work_sync() to sleep if irq_work() no IRQ support.
sched/rt: Annotate the RT balancing logic irqwork as IRQ_WORK_HARD_IRQ
sched: Add cluster scheduler level for x86
sched: Add cluster scheduler level in core and related Kconfig for ARM64
topology: Represent clusters of CPUs within a die
sched: Disable -Wunused-but-set-variable
sched: Add wrapper for get_wchan() to keep task blocked
x86: Fix get_wchan() to support the ORC unwinder
proc: Use task_is_running() for wchan in /proc/$pid/stat
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull objtool updates from Thomas Gleixner:
- Improve retpoline code patching by separating it from alternatives
which reduces memory footprint and allows to do better optimizations
in the actual runtime patching.
- Add proper retpoline support for x86/BPF
- Address noinstr warnings in x86/kvm, lockdep and paravirtualization
code
- Add support to handle pv_opsindirect calls in the noinstr analysis
- Classify symbols upfront and cache the result to avoid redundant
str*cmp() invocations.
- Add a CFI hash to reduce memory consumption which also reduces
runtime on a allyesconfig by ~50%
- Adjust XEN code to make objtool handling more robust and as a side
effect to prevent text fragmentation due to placement of the
hypercall page.
* tag 'objtool-core-2021-10-31' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (41 commits)
bpf,x86: Respect X86_FEATURE_RETPOLINE*
bpf,x86: Simplify computing label offsets
x86,bugs: Unconditionally allow spectre_v2=retpoline,amd
x86/alternative: Add debug prints to apply_retpolines()
x86/alternative: Try inline spectre_v2=retpoline,amd
x86/alternative: Handle Jcc __x86_indirect_thunk_\reg
x86/alternative: Implement .retpoline_sites support
x86/retpoline: Create a retpoline thunk array
x86/retpoline: Move the retpoline thunk declarations to nospec-branch.h
x86/asm: Fixup odd GEN-for-each-reg.h usage
x86/asm: Fix register order
x86/retpoline: Remove unused replacement symbols
objtool,x86: Replace alternatives with .retpoline_sites
objtool: Shrink struct instruction
objtool: Explicitly avoid self modifying code in .altinstr_replacement
objtool: Classify symbols
objtool: Support pv_opsindirect calls for noinstr
x86/xen: Rework the xen_{cpu,irq,mmu}_opsarrays
x86/xen: Mark xen_force_evtchn_callback() noinstr
x86/xen: Make irq_disable() noinstr
...
|
|
Currently Linux prevents usage of retpoline,amd on !AMD hardware, this
is unfriendly and gets in the way of testing. Remove this restriction.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Borislav Petkov <bp@suse.de>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Tested-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/r/20211026120310.487348118@infradead.org
|
|
Hyper-V exposes shared memory boundary via cpuid
HYPERV_CPUID_ISOLATION_CONFIG and store it in the
shared_gpa_boundary of ms_hyperv struct. This prepares
to share memory with host for SNP guest.
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Tianyu Lan <Tianyu.Lan@microsoft.com>
Link: https://lore.kernel.org/r/20211025122116.264793-3-ltykernel@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
Hyperv exposes GHCB page via SEV ES GHCB MSR for SNP guest
to communicate with hypervisor. Map GHCB page for all
cpus to read/write MSR register and submit hvcall request
via ghcb page.
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: Tianyu Lan <Tianyu.Lan@microsoft.com>
Link: https://lore.kernel.org/r/20211025122116.264793-2-ltykernel@gmail.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
|
|
The msm next tree is based on rc3, so let's just backmerge rc7 before pulling it in.
Signed-off-by: Dave Airlie <airlied@redhat.com>
|
|
Add the AMX state components in XFEATURE_MASK_USER_SUPPORTED and the
TILE_DATA component to the dynamic states and update the permission check
table accordingly.
This is only effective on 64 bit kernels as for 32bit kernels
XFEATURE_MASK_TILE is defined as 0.
TILE_DATA is caller-saved state and the only dynamic state. Add build time
sanity check to ensure the assumption that every dynamic feature is caller-
saved.
Make AMX state depend on XFD as it is dynamic feature.
Signed-off-by: Chang S. Bae <chang.seok.bae@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20211021225527.10184-24-chang.seok.bae@intel.com
|
|
Intel's eXtended Feature Disable (XFD) feature is an extension of the XSAVE
architecture. XFD allows the kernel to enable a feature state in XCR0 and
to receive a #NM trap when a task uses instructions accessing that state.
This is going to be used to postpone the allocation of a larger XSTATE
buffer for a task to the point where it is actually using a related
instruction after the permission to use that facility has been granted.
XFD is not used by the kernel, but only applied to userspace. This is a
matter of policy as the kernel knows how a fpstate is reallocated and the
XFD state.
The compacted XSAVE format is adjustable for dynamic features. Make XFD
depend on XSAVES.
Signed-off-by: Chang S. Bae <chang.seok.bae@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Chang S. Bae <chang.seok.bae@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20211021225527.10184-13-chang.seok.bae@intel.com
|
|
For bare-metal SGX on real hardware, the hardware provides guarantees
SGX state at reboot. For instance, all pages start out uninitialized.
The vepc driver provides a similar guarantee today for freshly-opened
vepc instances, but guests such as Windows expect all pages to be in
uninitialized state on startup, including after every guest reboot.
Some userspace implementations of virtual SGX would rather avoid having
to close and reopen the /dev/sgx_vepc file descriptor and re-mmap the
virtual EPC. For example, they could sandbox themselves after the guest
starts and forbid further calls to open(), in order to mitigate exploits
from untrusted guests.
Therefore, add a ioctl that does this with EREMOVE. Userspace can
invoke the ioctl to bring its vEPC pages back to uninitialized state.
There is a possibility that some pages fail to be removed if they are
SECS pages, and the child and SECS pages could be in separate vEPC
regions. Therefore, the ioctl returns the number of EREMOVE failures,
telling userspace to try the ioctl again after it's done with all
vEPC regions. A more verbose description of the correct usage and
the possible error conditions is documented in sgx.rst.
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: https://lkml.kernel.org/r/20211021201155.1523989-3-pbonzini@redhat.com
|
|
For bare-metal SGX on real hardware, the hardware provides guarantees
SGX state at reboot. For instance, all pages start out uninitialized.
The vepc driver provides a similar guarantee today for freshly-opened
vepc instances, but guests such as Windows expect all pages to be in
uninitialized state on startup, including after every guest reboot.
One way to do this is to simply close and reopen the /dev/sgx_vepc file
descriptor and re-mmap the virtual EPC. However, this is problematic
because it prevents sandboxing the userspace (for example forbidding
open() after the guest starts; this is doable with heavy use of SCM_RIGHTS
file descriptor passing).
In order to implement this, we will need a ioctl that performs
EREMOVE on all pages mapped by a /dev/sgx_vepc file descriptor:
other possibilities, such as closing and reopening the device,
are racy.
Start the implementation by creating a separate function with just
the __eremove wrapper.
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: https://lkml.kernel.org/r/20211021201155.1523989-2-pbonzini@redhat.com
|
|
The microcode loader has been looping through __start_builtin_fw down to
__end_builtin_fw to look for possibly built-in firmware for microcode
updates.
Now that the firmware loader code has exported an API for looping
through the kernel's built-in firmware section, use it and drop the x86
implementation in favor.
Signed-off-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Link: https://lore.kernel.org/r/20211021155843.1969401-4-mcgrof@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Currently, Linux probes for X86_BUG_NULL_SEL unconditionally which
makes it unsafe to migrate in a virtualised environment as the
properties across the migration pool might differ.
To be specific, the case which goes wrong is:
1. Zen1 (or earlier) and Zen2 (or later) in a migration pool
2. Linux boots on Zen2, probes and finds the absence of X86_BUG_NULL_SEL
3. Linux is then migrated to Zen1
Linux is now running on a X86_BUG_NULL_SEL-impacted CPU while believing
that the bug is fixed.
The only way to address the problem is to fully trust the "no longer
affected" CPUID bit when virtualised, because in the above case it would
be clear deliberately to indicate the fact "you might migrate to
somewhere which has this behaviour".
Zen3 adds the NullSelectorClearsBase CPUID bit to indicate that loading
a NULL segment selector zeroes the base and limit fields, as well as
just attributes. Zen2 also has this behaviour but doesn't have the NSCB
bit.
[ bp: Minor touchups. ]
Signed-off-by: Jane Malalane <jane.malalane@citrix.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
CC: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20211021104744.24126-1-jane.malalane@citrix.com
|
|
DM&P devices were not being properly identified, which resulted in
unneeded Spectre/Meltdown mitigations being applied.
The manufacturer states that these devices execute always in-order and
don't support either speculative execution or branch prediction, so
they are not vulnerable to this class of attack. [1]
This is something I've personally tested by a simple timing analysis
on my Vortex86MX CPU, and can confirm it is true.
Add identification for some devices that lack the CPUID product name
call, so they appear properly on /proc/cpuinfo.
¹https://www.ssv-embedded.de/doks/infos/DMP_Ann_180108_Meltdown.pdf
[ bp: Massage commit message. ]
Signed-off-by: Marcos Del Sol Vives <marcos@orca.pet>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20211017094408.1512158-1-marcos@orca.pet
|
|
Now that the file is empty, fixup all references with the proper includes
and delete the former kitchen sink.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20211015011540.001197214@linutronix.de
|
|
Resolve the conflict between these commits:
x86/fpu: 1193f408cd51 ("x86/fpu/signal: Change return type of __fpu_restore_sig() to boolean")
x86/urgent: d298b03506d3 ("x86/fpu: Restore the masking out of reserved MXCSR bits")
b2381acd3fd9 ("x86/fpu: Mask out the invalid MXCSR bits properly")
Conflicts:
arch/x86/kernel/fpu/signal.c
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
There are x86 CPU architectures (e.g. Jacobsville) where L2 cahce is
shared among a cluster of cores instead of being exclusive to one
single core.
To prevent oversubscription of L2 cache, load should be balanced
between such L2 clusters, especially for tasks with no shared data.
On benchmark such as SPECrate mcf test, this change provides a boost
to performance especially on medium load system on Jacobsville. on a
Jacobsville that has 24 Atom cores, arranged into 6 clusters of 4
cores each, the benchmark number is as follow:
Improvement over baseline kernel for mcf_r
copies run time base rate
1 -0.1% -0.2%
6 25.1% 25.1%
12 18.8% 19.0%
24 0.3% 0.3%
So this looks pretty good. In terms of the system's task distribution,
some pretty bad clumping can be seen for the vanilla kernel without
the L2 cluster domain for the 6 and 12 copies case. With the extra
domain for cluster, the load does get evened out between the clusters.
Note this patch isn't an universal win as spreading isn't necessarily
a win, particually for those workload who can benefit from packing.
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210924085104.44806-4-21cnbao@gmail.com
|
|
Export smca_get_bank_type for use in the AMD GPU
driver to determine MCA bank while handling correctable
and uncorrectable errors in GPU UMC.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
Commit
3c73b81a9164 ("x86/entry, selftests: Further improve user entry sanity checks")
added a warning if AC is set when in the kernel.
Commit
662a0221893a3d ("x86/entry: Fix AC assertion")
changed the warning to only fire if the CPU supports SMAP.
However, the warning can still trigger on a machine that supports SMAP
but where it's disabled in the kernel config and when running the
syscall_nt selftest, for example:
------------[ cut here ]------------
WARNING: CPU: 0 PID: 49 at irqentry_enter_from_user_mode
CPU: 0 PID: 49 Comm: init Tainted: G T 5.15.0-rc4+ #98 e6202628ee053b4f310759978284bd8bb0ce6905
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014
RIP: 0010:irqentry_enter_from_user_mode
...
Call Trace:
? irqentry_enter
? exc_general_protection
? asm_exc_general_protection
? asm_exc_general_protectio
IS_ENABLED(CONFIG_X86_SMAP) could be added to the warning condition, but
even this would not be enough in case SMAP is disabled at boot time with
the "nosmap" parameter.
To be consistent with "nosmap" behaviour, clear X86_FEATURE_SMAP when
!CONFIG_X86_SMAP.
Found using entry-fuzz + satrandconfig.
[ bp: Massage commit message. ]
Fixes: 3c73b81a9164 ("x86/entry, selftests: Further improve user entry sanity checks")
Fixes: 662a0221893a ("x86/entry: Fix AC assertion")
Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20211003223423.8666-1-vegard.nossum@oracle.com
|
|
Commit in Fixes separated the architecture specific and filesystem parts
of the resctrl domain structures.
This left the error paths in domain_add_cpu() kfree()ing the memory with
the wrong type.
This will cause a problem if someone adds a new member to struct
rdt_hw_domain meaning d_resctrl is no longer the first member.
Fixes: 792e0f6f789b ("x86/resctrl: Split struct rdt_domain")
Signed-off-by: James Morse <james.morse@arm.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Reinette Chatre <reinette.chatre@intel.com>
Link: https://lkml.kernel.org/r/20210917165924.28254-1-james.morse@arm.com
|
|
domain_add_cpu() is called whenever a CPU is brought online. The
earlier call to domain_setup_ctrlval() allocates the control value
arrays.
If domain_setup_mon_state() fails, the control value arrays are not
freed.
Add the missing kfree() calls.
Fixes: 1bd2a63b4f0de ("x86/intel_rdt/mba_sc: Add initialization support")
Fixes: edf6fa1c4a951 ("x86/intel_rdt/cqm: Add RMID (Resource monitoring ID) management")
Signed-off-by: James Morse <james.morse@arm.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Reinette Chatre <reinette.chatre@intel.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20210917165958.28313-1-james.morse@arm.com
|
|
Switch the kernel default of SSBD and STIBP to the ones with
CONFIG_SECCOMP=n (i.e. spec_store_bypass_disable=prctl
spectre_v2_user=prctl) even if CONFIG_SECCOMP=y.
Several motivations listed below:
- If SMT is enabled the seccomp jail can still attack the rest of the
system even with spectre_v2_user=seccomp by using MDS-HT (except on
XEON PHI where MDS can be tamed with SMT left enabled, but that's a
special case). Setting STIBP become a very expensive window dressing
after MDS-HT was discovered.
- The seccomp jail cannot attack the kernel with spectre-v2-HT
regardless (even if STIBP is not set), but with MDS-HT the seccomp
jail can attack the kernel too.
- With spec_store_bypass_disable=prctl the seccomp jail can attack the
other userland (guest or host mode) using spectre-v2-HT, but the
userland attack is already mitigated by both ASLR and pid namespaces
for host userland and through virt isolation with libkrun or
kata. (if something if somebody is worried about spectre-v2-HT it's
best to mount proc with hidepid=2,gid=proc on workstations where not
all apps may run under container runtimes, rather than slowing down
all seccomp jails, but the best is to add pid namespaces to the
seccomp jail). As opposed MDS-HT is not mitigated and the seccomp
jail can still attack all other host and guest userland if SMT is
enabled even with spec_store_bypass_disable=seccomp.
- If full security is required then MDS-HT must also be mitigated with
nosmt and then spectre_v2_user=prctl and spectre_v2_user=seccomp
would become identical.
- Setting spectre_v2_user=seccomp is overall lower priority than to
setting javascript.options.wasm false in about:config to protect
against remote wasm MDS-HT, instead of worrying about Spectre-v2-HT
and STIBP which again is already statistically well mitigated by
other means in userland and it's fully mitigated in kernel with
retpolines (unlike the wasm assist call with MDS-HT).
- SSBD is needed to prevent reading the JIT memory and the primary
user being the OpenJDK. However the primary user of SSBD wouldn't be
covered by spec_store_bypass_disable=seccomp because it doesn't use
seccomp and the primary user also explicitly declined to set
PR_SET_SPECULATION_CTRL+PR_SPEC_STORE_BYPASS despite it easily
could. In fact it would need to set it only when the sandboxing
mechanism is enabled for javaws applets, but it still declined it by
declaring security within the same user address space as an
untenable objective for their JIT, even in the sandboxing case where
performance would be a lesser concern (for the record: I kind of
disagree in not setting PR_SPEC_STORE_BYPASS in the sandbox case and
I prefer to run javaws through a wrapper that sets
PR_SPEC_STORE_BYPASS if I need). In turn it can be inferred that
even if the primary user of SSBD would use seccomp, they would
invoke it with SECCOMP_FILTER_FLAG_SPEC_ALLOW by now.
- runc/crun already set SECCOMP_FILTER_FLAG_SPEC_ALLOW by default, k8s
and podman have a default json seccomp allowlist that cannot be
slowed down, so for the #1 seccomp user this change is already a
noop.
- systemd/sshd or other apps that use seccomp, if they really need
STIBP or SSBD, they need to explicitly set the
PR_SET_SPECULATION_CTRL by now. The stibp/ssbd seccomp blind
catch-all approach was done probably initially with a wishful
thinking objective to pretend to have a peace of mind that it could
magically fix it all. That was wishful thinking before MDS-HT was
discovered, but after MDS-HT has been discovered it become just
window dressing.
- For qemu "-sandbox" seccomp jail it wouldn't make sense to set STIBP
or SSBD. SSBD doesn't help with KVM because there's no JIT (if it's
needed with TCG it should be an opt-in with
PR_SET_SPECULATION_CTRL+PR_SPEC_STORE_BYPASS and it shouldn't
slowdown KVM for nothing). For qemu+KVM STIBP would be even more
window dressing than it is for all other apps, because in the
qemu+KVM case there's not only the MDS attack to worry about with
SMT enabled. Even after disabling SMT, there's still a theoretical
spectre-v2 attack possible within the same thread context from guest
mode to host ring3 that the host kernel retpoline mitigation has no
theoretical chance to mitigate. On some kernels a
ibrs-always/ibrs-retpoline opt-in model is provided that will
enabled IBRS in the qemu host ring3 userland which fixes this
theoretical concern. Only after enabling IBRS in the host userland
it would then make sense to proceed and worry about STIBP and an
attack on the other host userland, but then again SMT would need to
be disabled for full security anyway, so that would render STIBP
again a noop.
- last but not the least: the lack of "spec_store_bypass_disable=prctl
spectre_v2_user=prctl" means the moment a guest boots and
sshd/systemd runs, the guest kernel will write to SPEC_CTRL MSR
which will make the guest vmexit forever slower, forcing KVM to
issue a very slow rdmsr instruction at every vmexit. So the end
result is that SPEC_CTRL MSR is only available in GCE. Most other
public cloud providers don't expose SPEC_CTRL, which means that not
only STIBP/SSBD isn't available, but IBPB isn't available either
(which would cause no overhead to the guest or the hypervisor
because it's write only and requires no reading during vmexit). So
the current default already net loss in security (missing IBPB)
which means most public cloud providers cannot achieve a fully
secure guest with nosmt (and nosmt is enough to fully mitigate
MDS-HT). It also means GCE and is unfairly penalized in performance
because it provides the option to enable full security in the guest
as an opt-in (i.e. nosmt and IBPB). So this change will allow all
cloud providers to expose SPEC_CTRL without incurring into any
hypervisor slowdown and at the same time it will remove the unfair
penalization of GCE performance for doing the right thing and it'll
allow to get full security with nosmt with IBPB being available (and
STIBP becoming meaningless).
Example to put things in prospective: the STIBP enabled in seccomp has
never been about protecting apps using seccomp like sshd from an
attack from a malicious userland, but to the contrary it has always
been about protecting the system from an attack from sshd, after a
successful remote network exploit against sshd. In fact initially it
wasn't obvious STIBP would work both ways (STIBP was about preventing
the task that runs with STIBP to be attacked with spectre-v2-HT, but
accidentally in the STIBP case it also prevents the attack in the
other direction). In the hypothetical case that sshd has been remotely
exploited the last concern should be STIBP being set, because it'll be
still possible to obtain info even from the kernel by using MDS if
nosmt wasn't set (and if it was set, STIBP is a noop in the first
place). As opposed kernel cannot leak anything with spectre-v2 HT
because of retpolines and the userland is mitigated by ASLR already
and ideally PID namespaces too. If something it'd be worth checking if
sshd run the seccomp thread under pid namespaces too if available in
the running kernel. SSBD also would be a noop for sshd, since sshd
uses no JIT. If sshd prefers to keep doing the STIBP window dressing
exercise, it still can even after this change of defaults by opting-in
with PR_SPEC_INDIRECT_BRANCH.
Ultimately setting SSBD and STIBP by default for all seccomp jails is
a bad sweet spot and bad default with more cons than pros that end up
reducing security in the public cloud (by giving an huge incentive to
not expose SPEC_CTRL which would be needed to get full security with
IBPB after setting nosmt in the guest) and by excessively hurting
performance to more secure apps using seccomp that end up having to
opt out with SECCOMP_FILTER_FLAG_SPEC_ALLOW.
The following is the verified result of the new default with SMT
enabled:
(gdb) print spectre_v2_user_stibp
$1 = SPECTRE_V2_USER_PRCTL
(gdb) print spectre_v2_user_ibpb
$2 = SPECTRE_V2_USER_PRCTL
(gdb) print ssb_mode
$3 = SPEC_STORE_BYPASS_PRCTL
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20201104235054.5678-1-aarcange@redhat.com
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Link: https://lore.kernel.org/lkml/AAA2EF2C-293D-4D5B-BFA6-FF655105CD84@redhat.com
Acked-by: Waiman Long <longman@redhat.com>
Link: https://lore.kernel.org/lkml/c0722838-06f7-da6b-138f-e0f26362f16a@redhat.com
|
|
$ pahole -C mca_config arch/x86/kernel/cpu/mce/core.o
before:
/* size: 40, cachelines: 1, members: 16 */
/* sum members: 21, holes: 1, sum holes: 3 */
/* sum bitfield members: 64 bits, bit holes: 2, sum bit holes: 32 bits */
/* padding: 4 */
/* last cacheline: 40 bytes */
after:
/* size: 32, cachelines: 1, members: 16 */
/* padding: 3 */
/* last cacheline: 32 bytes */
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20210922165101.18951-6-bp@alien8.de
|
|
Use a flag setting to call the only quirk function for that.
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20210922165101.18951-5-bp@alien8.de
|
|
Avoid having indirect calls and use a normal function which returns the
proper MSR address based on ->smca setting.
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20210922165101.18951-4-bp@alien8.de
|
|
Get rid of the indirect function pointer and use flags settings instead
to steer execution.
Now that it is not an indirect call any longer, drop the instrumentation
annotation for objtool too.
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20210922165101.18951-3-bp@alien8.de
|
|
Turn it into a normal function which calls an AMD- or Intel-specific
variant depending on the CPU it runs on.
No functional changes.
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20210922165101.18951-2-bp@alien8.de
|
|
Sending a SIGBUS for a copy from user is not the correct semantic.
System calls should return -EFAULT (or a short count for write(2)).
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210818002942.1607544-3-tony.luck@intel.com
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
- Prevent a infinite loop in the MCE recovery on return to user space,
which was caused by a second MCE queueing work for the same page and
thereby creating a circular work list.
- Make kern_addr_valid() handle existing PMD entries, which are marked
not present in the higher level page table, correctly instead of
blindly dereferencing them.
- Pass a valid address to sanitize_phys(). This was caused by the
mixture of inclusive and exclusive ranges. memtype_reserve() expect
'end' being exclusive, but sanitize_phys() wants it inclusive. This
worked so far, but with end being the end of the physical address
space the fail is exposed.
- Increase the maximum supported GPIO numbers for 64bit. Newer SoCs
exceed the previous maximum.
* tag 'x86_urgent_for_v5.15_rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mce: Avoid infinite loop for copy from user recovery
x86/mm: Fix kern_addr_valid() to cope with existing but not present entries
x86/platform: Increase maximum GPIO number for X86_64
x86/pat: Pass valid address to sanitize_phys()
|
|
There are two cases for machine check recovery:
1) The machine check was triggered by ring3 (application) code.
This is the simpler case. The machine check handler simply queues
work to be executed on return to user. That code unmaps the page
from all users and arranges to send a SIGBUS to the task that
triggered the poison.
2) The machine check was triggered in kernel code that is covered by
an exception table entry. In this case the machine check handler
still queues a work entry to unmap the page, etc. but this will
not be called right away because the #MC handler returns to the
fix up code address in the exception table entry.
Problems occur if the kernel triggers another machine check before the
return to user processes the first queued work item.
Specifically, the work is queued using the ->mce_kill_me callback
structure in the task struct for the current thread. Attempting to queue
a second work item using this same callback results in a loop in the
linked list of work functions to call. So when the kernel does return to
user, it enters an infinite loop processing the same entry for ever.
There are some legitimate scenarios where the kernel may take a second
machine check before returning to the user.
1) Some code (e.g. futex) first tries a get_user() with page faults
disabled. If this fails, the code retries with page faults enabled
expecting that this will resolve the page fault.
2) Copy from user code retries a copy in byte-at-time mode to check
whether any additional bytes can be copied.
On the other side of the fence are some bad drivers that do not check
the return value from individual get_user() calls and may access
multiple user addresses without noticing that some/all calls have
failed.
Fix by adding a counter (current->mce_count) to keep track of repeated
machine checks before task_work() is called. First machine check saves
the address information and calls task_work_add(). Subsequent machine
checks before that task_work call back is executed check that the address
is in the same page as the first machine check (since the callback will
offline exactly one page).
Expected worst case is four machine checks before moving on (e.g. one
user access with page faults disabled, then a repeat to the same address
with page faults enabled ... repeat in copy tail bytes). Just in case
there is some code that loops forever enforce a limit of 10.
[ bp: Massage commit message, drop noinstr, fix typo, extend panic
messages. ]
Fixes: 5567d11c21a1 ("x86/mce: Send #MC singal from task work")
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/YT/IJ9ziLqmtqEPu@agluck-desk2.amr.corp.intel.com
|
|
Now that the MC safe copy and FPU have been converted to use the MCE safe
fixup types remove EX_TYPE_FAULT from the list of types which MCE considers
to be safe to be recovered in kernel.
This removes the SGX exception handling of ENCLS from the #MC safe
handling, but according to the SGX wizards the current SGX implementations
cannot survive #MC on ENCLS:
https://lore.kernel.org/r/YS+upEmTfpZub3s9@google.com
The code relies on the trap number being stored if ENCLS raised an
exception. That's still working, but it does no longer trick the MCE code
into assuming that #MC is handled correctly for ENCLS.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210908132525.445255957@linutronix.de
|
|
Provide exception fixup types which can be used to identify fixups which
allow in kernel #MC recovery and make them invoke the existing handlers.
These will be used at places where #MC recovery is handled correctly by the
caller.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210908132525.269689153@linutronix.de
|
|
The exception table entries contain the instruction address, the fixup
address and the handler address. All addresses are relative. Storing the
handler address has a few downsides:
1) Most handlers need to be exported
2) Handlers can be defined everywhere and there is no overview about the
handler types
3) MCE needs to check the handler type to decide whether an in kernel #MC
can be recovered. The functionality of the handler itself is not in any
way special, but for these checks there need to be separate functions
which in the worst case have to be exported.
Some of these 'recoverable' exception fixups are pretty obscure and
just reuse some other handler to spare code. That obfuscates e.g. the
#MC safe copy functions. Cleaning that up would require more handlers
and exports
Rework the exception fixup mechanics by storing a fixup type number instead
of the handler address and invoke the proper handler for each fixup
type. Also teach the extable sort to leave the type field alone.
This makes most handlers static except for special cases like the MCE
MSR fixup and the BPF fixup. This allows to add more types for cleaning up
the obscure places without adding more handler code and exports.
There is a marginal code size reduction for a production config and it
removes _eight_ exported symbols.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lkml.kernel.org/r/20210908132525.211958725@linutronix.de
|
|
and the random number of tabs.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210908132525.154428878@linutronix.de
|
|
Prepare code for further simplification. No functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210908132525.096452100@linutronix.de
|
|
Ensure that all usage sites of get/put_online_cpus() except for the
struggler in drivers/thermal are gone. So the last user and the deprecated
inlines can be removed.
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux
Pull hyperv updates from Wei Liu:
- make Hyper-V code arch-agnostic (Michael Kelley)
- fix sched_clock behaviour on Hyper-V (Ani Sinha)
- fix a fault when Linux runs as the root partition on MSHV (Praveen
Kumar)
- fix VSS driver (Vitaly Kuznetsov)
- cleanup (Sonia Sharma)
* tag 'hyperv-next-signed-20210831' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
hv_utils: Set the maximum packet size for VSS driver to the length of the receive buffer
Drivers: hv: Enable Hyper-V code to be built on ARM64
arm64: efi: Export screen_info
arm64: hyperv: Initialize hypervisor on boot
arm64: hyperv: Add panic handler
arm64: hyperv: Add Hyper-V hypercall and register access utilities
x86/hyperv: fix root partition faults when writing to VP assist page MSR
hv: hyperv.h: Remove unused inline functions
drivers: hv: Decouple Hyper-V clock/timer code from VMbus drivers
x86/hyperv: add comment describing TSC_INVARIANT_CONTROL MSR setting bit 0
Drivers: hv: Move Hyper-V misc functionality to arch-neutral code
Drivers: hv: Add arch independent default functions for some Hyper-V handlers
Drivers: hv: Make portions of Hyper-V init code be arch neutral
x86/hyperv: fix for unwanted manipulation of sched_clock when TSC marked unstable
asm-generic/hyperv: Add missing #include of nmi.h
|
