When the AMD QoS feature CDP (code and data prioritization) is enabled
or disabled, the CDP bit in MSR 0000_0C81 is written on one of the CPUs
in an L3 domain (core complex). That is not correct - the CDP bit needs
to be updated on all the logical CPUs in the domain.
This was not spelled out clearly in the spec earlier. The specification
has been updated and the updated document, "AMD64 Technology Platform
Quality of Service Extensions Publication # 56375 Revision: 1.02 Issue
Date: October 2020" is available now. Refer the section: Code and Data
Prioritization.
Fix the issue by adding a new flag arch_has_per_cpu_cfg in rdt_cache
data structure.
The documentation can be obtained at:
https://developer.amd.com/wp-content/resources/56375.pdf
Link: https://bugzilla.kernel.org/show_bug.cgi?id=206537
[ bp: Massage commit message. ]
Fixes: 4d05bf71f1 ("x86/resctrl: Introduce AMD QOS feature")
Signed-off-by: Babu Moger <babu.moger@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Link: https://lkml.kernel.org/r/160675180380.15628.3309402017215002347.stgit@bmoger-ubuntu
Currently, if mce_end() fails, no_way_out - the variable denoting
whether the machine can recover from this MCE - is determined by whether
the worst severity that was found across the MCA banks associated with
the current CPU, is of panic severity.
However, at this point no_way_out could have been already set by
mca_start() after looking at all severities of all CPUs that entered the
MCE handler. If mce_end() fails, check first if no_way_out is already
set and, if so, stick to it, otherwise use the local worst value.
[ bp: Massage. ]
Signed-off-by: Gabriele Paoloni <gabriele.paoloni@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20201127161819.3106432-2-gabriele.paoloni@intel.com
When spectre_v2_user={seccomp,prctl},ibpb is specified on the command
line, IBPB is force-enabled and STIPB is conditionally-enabled (or not
available).
However, since
21998a3515 ("x86/speculation: Avoid force-disabling IBPB based on STIBP and enhanced IBRS.")
the spectre_v2_user_ibpb variable is set to SPECTRE_V2_USER_{PRCTL,SECCOMP}
instead of SPECTRE_V2_USER_STRICT, which is the actual behaviour.
Because the issuing of IBPB relies on the switch_mm_*_ibpb static
branches, the mitigations behave as expected.
Since
1978b3a53a ("x86/speculation: Allow IBPB to be conditionally enabled on CPUs with always-on STIBP")
this discrepency caused the misreporting of IB speculation via prctl().
On CPUs with STIBP always-on and spectre_v2_user=seccomp,ibpb,
prctl(PR_GET_SPECULATION_CTRL) would return PR_SPEC_PRCTL |
PR_SPEC_ENABLE instead of PR_SPEC_DISABLE since both IBPB and STIPB are
always on. It also allowed prctl(PR_SET_SPECULATION_CTRL) to set the IB
speculation mode, even though the flag is ignored.
Similarly, for CPUs without SMT, prctl(PR_GET_SPECULATION_CTRL) should
also return PR_SPEC_DISABLE since IBPB is always on and STIBP is not
available.
[ bp: Massage commit message. ]
Fixes: 21998a3515 ("x86/speculation: Avoid force-disabling IBPB based on STIBP and enhanced IBRS.")
Fixes: 1978b3a53a ("x86/speculation: Allow IBPB to be conditionally enabled on CPUs with always-on STIBP")
Signed-off-by: Anand K Mistry <amistry@google.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20201110123349.1.Id0cbf996d2151f4c143c90f9028651a5b49a5908@changeid
On resource group creation via a mkdir an extra kernfs_node reference is
obtained by kernfs_get() to ensure that the rdtgroup structure remains
accessible for the rdtgroup_kn_unlock() calls where it is removed on
deletion. Currently the extra kernfs_node reference count is only
dropped by kernfs_put() in rdtgroup_kn_unlock() while the rdtgroup
structure is removed in a few other locations that lack the matching
reference drop.
In call paths of rmdir and umount, when a control group is removed,
kernfs_remove() is called to remove the whole kernfs nodes tree of the
control group (including the kernfs nodes trees of all child monitoring
groups), and then rdtgroup structure is freed by kfree(). The rdtgroup
structures of all child monitoring groups under the control group are
freed by kfree() in free_all_child_rdtgrp().
Before calling kfree() to free the rdtgroup structures, the kernfs node
of the control group itself as well as the kernfs nodes of all child
monitoring groups still take the extra references which will never be
dropped to 0 and the kernfs nodes will never be freed. It leads to
reference count leak and kernfs_node_cache memory leak.
For example, reference count leak is observed in these two cases:
(1) mount -t resctrl resctrl /sys/fs/resctrl
mkdir /sys/fs/resctrl/c1
mkdir /sys/fs/resctrl/c1/mon_groups/m1
umount /sys/fs/resctrl
(2) mkdir /sys/fs/resctrl/c1
mkdir /sys/fs/resctrl/c1/mon_groups/m1
rmdir /sys/fs/resctrl/c1
The same reference count leak issue also exists in the error exit paths
of mkdir in mkdir_rdt_prepare() and rdtgroup_mkdir_ctrl_mon().
Fix this issue by following changes to make sure the extra kernfs_node
reference on rdtgroup is dropped before freeing the rdtgroup structure.
(1) Introduce rdtgroup removal helper rdtgroup_remove() to wrap up
kernfs_put() and kfree().
(2) Call rdtgroup_remove() in rdtgroup removal path where the rdtgroup
structure is about to be freed by kfree().
(3) Call rdtgroup_remove() or kernfs_put() as appropriate in the error
exit paths of mkdir where an extra reference is taken by kernfs_get().
Fixes: f3cbeacaa0 ("x86/intel_rdt/cqm: Add rmdir support")
Fixes: e02737d5b8 ("x86/intel_rdt: Add tasks files")
Fixes: 60cf5e101f ("x86/intel_rdt: Add mkdir to resctrl file system")
Reported-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1604085088-31707-1-git-send-email-xiaochen.shen@intel.com
Willem reported growing of kernfs_node_cache entries in slabtop when
repeatedly creating and removing resctrl subdirectories as well as when
repeatedly mounting and unmounting the resctrl filesystem.
On resource group (control as well as monitoring) creation via a mkdir
an extra kernfs_node reference is obtained to ensure that the rdtgroup
structure remains accessible for the rdtgroup_kn_unlock() calls where it
is removed on deletion. The kernfs_node reference count is dropped by
kernfs_put() in rdtgroup_kn_unlock().
With the above explaining the need for one kernfs_get()/kernfs_put()
pair in resctrl there are more places where a kernfs_node reference is
obtained without a corresponding release. The excessive amount of
reference count on kernfs nodes will never be dropped to 0 and the
kernfs nodes will never be freed in the call paths of rmdir and umount.
It leads to reference count leak and kernfs_node_cache memory leak.
Remove the superfluous kernfs_get() calls and expand the existing
comments surrounding the remaining kernfs_get()/kernfs_put() pair that
remains in use.
Superfluous kernfs_get() calls are removed from two areas:
(1) In call paths of mount and mkdir, when kernfs nodes for "info",
"mon_groups" and "mon_data" directories and sub-directories are
created, the reference count of newly created kernfs node is set to 1.
But after kernfs_create_dir() returns, superfluous kernfs_get() are
called to take an additional reference.
(2) kernfs_get() calls in rmdir call paths.
Fixes: 17eafd0762 ("x86/intel_rdt: Split resource group removal in two")
Fixes: 4af4a88e0c ("x86/intel_rdt/cqm: Add mount,umount support")
Fixes: f3cbeacaa0 ("x86/intel_rdt/cqm: Add rmdir support")
Fixes: d89b737901 ("x86/intel_rdt/cqm: Add mon_data")
Fixes: c7d9aac613 ("x86/intel_rdt/cqm: Add mkdir support for RDT monitoring")
Fixes: 5dc1d5c6ba ("x86/intel_rdt: Simplify info and base file lists")
Fixes: 60cf5e101f ("x86/intel_rdt: Add mkdir to resctrl file system")
Fixes: 4e978d06de ("x86/intel_rdt: Add "info" files to resctrl file system")
Reported-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Willem de Bruijn <willemb@google.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1604085053-31639-1-git-send-email-xiaochen.shen@intel.com
Fix
./arch/x86/kernel/cpu/sgx/ioctl.c:666: warning: Function parameter or member \
'encl' not described in 'sgx_ioc_enclave_provision'
./arch/x86/kernel/cpu/sgx/ioctl.c:666: warning: Excess function parameter \
'enclave' description in 'sgx_ioc_enclave_provision'
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201123181922.0c009406@canb.auug.org.au
The kernel uses ACPI Boot Error Record Table (BERT) to report fatal
errors that occurred in a previous boot. The MCA errors in the BERT are
reported using the x86 Processor Error Common Platform Error Record
(CPER) format. Currently, the record prints out the raw MSR values and
AMD relies on the raw record to provide MCA information.
Extract the raw MSR values of MCA registers from the BERT and feed them
into mce_log() to decode them properly.
The implementation is SMCA-specific as the raw MCA register values are
given in the register offset order of the SMCA address space.
[ bp: Massage. ]
[ Fix a build breakage in patch v1. ]
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lkml.kernel.org/r/20201119182938.151155-1-Smita.KoralahalliChannabasappa@amd.com
The call to rcu_cpu_starting() in mtrr_ap_init() is not early enough
in the CPU-hotplug onlining process, which results in lockdep splats
as follows:
=============================
WARNING: suspicious RCU usage
5.9.0+ #268 Not tainted
-----------------------------
kernel/kprobes.c:300 RCU-list traversed in non-reader section!!
other info that might help us debug this:
RCU used illegally from offline CPU!
rcu_scheduler_active = 1, debug_locks = 1
no locks held by swapper/1/0.
stack backtrace:
CPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.9.0+ #268
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-1ubuntu1 04/01/2014
Call Trace:
dump_stack+0x77/0x97
__is_insn_slot_addr+0x15d/0x170
kernel_text_address+0xba/0xe0
? get_stack_info+0x22/0xa0
__kernel_text_address+0x9/0x30
show_trace_log_lvl+0x17d/0x380
? dump_stack+0x77/0x97
dump_stack+0x77/0x97
__lock_acquire+0xdf7/0x1bf0
lock_acquire+0x258/0x3d0
? vprintk_emit+0x6d/0x2c0
_raw_spin_lock+0x27/0x40
? vprintk_emit+0x6d/0x2c0
vprintk_emit+0x6d/0x2c0
printk+0x4d/0x69
start_secondary+0x1c/0x100
secondary_startup_64_no_verify+0xb8/0xbb
This is avoided by moving the call to rcu_cpu_starting up near
the beginning of the start_secondary() function. Note that the
raw_smp_processor_id() is required in order to avoid calling into lockdep
before RCU has declared the CPU to be watched for readers.
Link: https://lore.kernel.org/lkml/160223032121.7002.1269740091547117869.tip-bot2@tip-bot2/
Reported-by: Qian Cai <cai@redhat.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The only usage of the kf_ops field in the rftype struct is to pass
it as argument to __kernfs_create_file(), which accepts a pointer to
const. Make it a pointer to const. This makes it possible to make
rdtgroup_kf_single_ops and kf_mondata_ops const, which allows the
compiler to put them in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Reinette Chatre <reinette.chatre@intel.com>
Link: https://lkml.kernel.org/r/20201110230228.801785-1-rikard.falkeborn@gmail.com
CPUID Leaf 0x1F defines a DIE_TYPE level (nb: ECX[8:15] level type == 0x5),
but CPUID Leaf 0xB does not. However, detect_extended_topology() will
set struct cpuinfo_x86.cpu_die_id regardless of whether a valid Die ID
was found.
Only set cpu_die_id if a DIE_TYPE level is found. CPU topology code may
use another value for cpu_die_id, e.g. the AMD NodeId on AMD-based
systems. Code ordering should be maintained so that the CPUID Leaf 0x1F
Die ID value will take precedence on systems that may use another value.
Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201109210659.754018-5-Yazen.Ghannam@amd.com
The Last Level Cache ID is returned by amd_get_nb_id(). In practice,
this value is the same as the AMD NodeId for callers of this function.
The NodeId is saved in struct cpuinfo_x86.cpu_die_id.
Replace calls to amd_get_nb_id() with the logical CPU's cpu_die_id and
remove the function.
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201109210659.754018-3-Yazen.Ghannam@amd.com
AMD systems provide a "NodeId" value that represents a global ID
indicating to which "Node" a logical CPU belongs. The "Node" is a
physical structure equivalent to a Die, and it should not be confused
with logical structures like NUMA nodes. Logical nodes can be adjusted
based on firmware or other settings whereas the physical nodes/dies are
fixed based on hardware topology.
The NodeId value can be used when a physical ID is needed by software.
Save the AMD NodeId to struct cpuinfo_x86.cpu_die_id. Use the value
from CPUID or MSR as appropriate. Default to phys_proc_id otherwise.
Do so for both AMD and Hygon systems.
Drop the node_id parameter from cacheinfo_*_init_llc_id() as it is no
longer needed.
Update the x86 topology documentation.
Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201109210659.754018-2-Yazen.Ghannam@amd.com
Short Version:
The SGX section->laundry_list structure is effectively thread-local, but
declared next to some shared structures. Its semantics are clear as mud.
Fix that. No functional changes. Compile tested only.
Long Version:
The SGX hardware keeps per-page metadata. This can provide things like
permissions, integrity and replay protection. It also prevents things
like having an enclave page mapped multiple times or shared between
enclaves.
But, that presents a problem for kexec()'d kernels (or any other kernel
that does not run immediately after a hardware reset). This is because
the last kernel may have been rude and forgotten to reset pages, which
would trigger the "shared page" sanity check.
To fix this, the SGX code "launders" the pages by running the EREMOVE
instruction on all pages at boot. This is slow and can take a long
time, so it is performed off in the SGX-specific ksgxd instead of being
synchronous at boot. The init code hands the list of pages to launder in
a per-SGX-section list: ->laundry_list. The only code to touch this list
is the init code and ksgxd. This means that no locking is necessary for
->laundry_list.
However, a lock is required for section->page_list, which is accessed
while creating enclaves and by ksgxd. This lock (section->lock) is
acquired by ksgxd while also processing ->laundry_list. It is easy to
confuse the purpose of the locking as being for ->laundry_list and
->page_list.
Rename ->laundry_list to ->init_laundry_list to make it clear that this
is not normally used at runtime. Also add some comments clarifying the
locking, and reorganize 'sgx_epc_section' to put 'lock' near the things
it protects.
Note: init_laundry_list is 128 bytes of wasted space at runtime. It
could theoretically be dynamically allocated and then freed after
the laundering process. But it would take nearly 128 bytes of extra
instructions to do that.
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201116222531.4834-1-dave.hansen@intel.com
Enclave memory is normally inaccessible from outside the enclave. This
makes enclaves hard to debug. However, enclaves can be put in a debug
mode when they are being built. In that mode, enclave data *can* be read
and/or written by using the ENCLS[EDBGRD] and ENCLS[EDBGWR] functions.
This is obviously only for debugging and destroys all the protections
present with normal enclaves. But, enclaves know their own debug status
and can adjust their behavior appropriately.
Add a vm_ops->access() implementation which can be used to read and write
memory inside debug enclaves. This is typically used via ptrace() APIs.
[ bp: Massage. ]
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-23-jarkko@kernel.org
Just like normal RAM, there is a limited amount of enclave memory available
and overcommitting it is a very valuable tool to reduce resource use.
Introduce a simple reclaim mechanism for enclave pages.
In contrast to normal page reclaim, the kernel cannot directly access
enclave memory. To get around this, the SGX architecture provides a set of
functions to help. Among other things, these functions copy enclave memory
to and from normal memory, encrypting it and protecting its integrity in
the process.
Implement a page reclaimer by using these functions. Picks victim pages in
LRU fashion from all the enclaves running in the system. A new kernel
thread (ksgxswapd) reclaims pages in the background based on watermarks,
similar to normal kswapd.
All enclave pages can be reclaimed, architecturally. But, there are some
limits to this, such as the special SECS metadata page which must be
reclaimed last. The page version array (used to mitigate replaying old
reclaimed pages) is also architecturally reclaimable, but not yet
implemented. The end result is that the vast majority of enclave pages are
currently reclaimable.
Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-22-jarkko@kernel.org
The whole point of SGX is to create a hardware protected place to do
“stuff”. But, before someone is willing to hand over the keys to
the castle , an enclave must often prove that it is running on an
SGX-protected processor. Provisioning enclaves play a key role in
providing proof.
There are actually three different enclaves in play in order to make this
happen:
1. The application enclave. The familiar one we know and love that runs
the actual code that’s doing real work. There can be many of these on
a single system, or even in a single application.
2. The quoting enclave (QE). The QE is mentioned in lots of silly
whitepapers, but, for the purposes of kernel enabling, just pretend they
do not exist.
3. The provisioning enclave. There is typically only one of these
enclaves per system. Provisioning enclaves have access to a special
hardware key.
They can use this key to help to generate certificates which serve as
proof that enclaves are running on trusted SGX hardware. These
certificates can be passed around without revealing the special key.
Any user who can create a provisioning enclave can access the
processor-unique Provisioning Certificate Key which has privacy and
fingerprinting implications. Even if a user is permitted to create
normal application enclaves (via /dev/sgx_enclave), they should not be
able to create provisioning enclaves. That means a separate permissions
scheme is needed to control provisioning enclave privileges.
Implement a separate device file (/dev/sgx_provision) which allows
creating provisioning enclaves. This device will typically have more
strict permissions than the plain enclave device.
The actual device “driver” is an empty stub. Open file descriptors for
this device will represent a token which allows provisioning enclave duty.
This file descriptor can be passed around and ultimately given as an
argument to the /dev/sgx_enclave driver ioctl().
[ bp: Touchups. ]
Suggested-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: linux-security-module@vger.kernel.org
Link: https://lkml.kernel.org/r/20201112220135.165028-16-jarkko@kernel.org
Enclaves have two basic states. They are either being built and are
malleable and can be modified by doing things like adding pages. Or,
they are locked down and not accepting changes. They can only be run
after they have been locked down. The ENCLS[EINIT] function induces the
transition from being malleable to locked-down.
Add an ioctl() that performs ENCLS[EINIT]. After this, new pages can
no longer be added with ENCLS[EADD]. This is also the time where the
enclave can be measured to verify its integrity.
Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-15-jarkko@kernel.org
SGX enclave pages are inaccessible to normal software. They must be
populated with data by copying from normal memory with the help of the
EADD and EEXTEND functions of the ENCLS instruction.
Add an ioctl() which performs EADD that adds new data to an enclave, and
optionally EEXTEND functions that hash the page contents and use the
hash as part of enclave “measurement” to ensure enclave integrity.
The enclave author gets to decide which pages will be included in the
enclave measurement with EEXTEND. Measurement is very slow and has
sometimes has very little value. For instance, an enclave _could_
measure every page of data and code, but would be slow to initialize.
Or, it might just measure its code and then trust that code to
initialize the bulk of its data after it starts running.
Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-14-jarkko@kernel.org
Intel(R) SGX is a new hardware functionality that can be used by
applications to set aside private regions of code and data called
enclaves. New hardware protects enclave code and data from outside
access and modification.
Add a driver that presents a device file and ioctl API to build and
manage enclaves.
[ bp: Small touchups, remove unused encl variable in sgx_encl_find() as
Reported-by: kernel test robot <lkp@intel.com> ]
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-12-jarkko@kernel.org
Add functions for runtime allocation and free.
This allocator and its algorithms are as simple as it gets. They do a
linear search across all EPC sections and find the first free page. They
are not NUMA-aware and only hand out individual pages. The SGX hardware
does not support large pages, so something more complicated like a buddy
allocator is unwarranted.
The free function (sgx_free_epc_page()) implicitly calls ENCLS[EREMOVE],
which returns the page to the uninitialized state. This ensures that the
page is ready for use at the next allocation.
Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-10-jarkko@kernel.org
Kernel support for SGX is ultimately decided by the state of the launch
control bits in the feature control MSR (MSR_IA32_FEAT_CTL). If the
hardware supports SGX, but neglects to support flexible launch control, the
kernel will not enable SGX.
Enable SGX at feature control MSR initialization and update the associated
X86_FEATURE flags accordingly. Disable X86_FEATURE_SGX (and all
derivatives) if the kernel is not able to establish itself as the authority
over SGX Launch Control.
All checks are performed for each logical CPU (not just boot CPU) in order
to verify that MSR_IA32_FEATURE_CONTROL is correctly configured on all
CPUs. All SGX code in this series expects the same configuration from all
CPUs.
This differs from VMX where X86_FEATURE_VMX is intentionally cleared only
for the current CPU so that KVM can provide additional information if KVM
fails to load like which CPU doesn't support VMX. There’s not much the
kernel or an administrator can do to fix the situation, so SGX neglects to
convey additional details about these kinds of failures if they occur.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Co-developed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-8-jarkko@kernel.org
Although carved out of normal DRAM, enclave memory is marked in the
system memory map as reserved and is not managed by the core mm. There
may be several regions spread across the system. Each contiguous region
is called an Enclave Page Cache (EPC) section. EPC sections are
enumerated via CPUID
Enclave pages can only be accessed when they are mapped as part of an
enclave, by a hardware thread running inside the enclave.
Parse CPUID data, create metadata for EPC pages and populate a simple
EPC page allocator. Although much smaller, ‘struct sgx_epc_page’
metadata is the SGX analog of the core mm ‘struct page’.
Similar to how the core mm’s page->flags encode zone and NUMA
information, embed the EPC section index to the first eight bits of
sgx_epc_page->desc. This allows a quick reverse lookup from EPC page to
EPC section. Existing client hardware supports only a single section,
while upcoming server hardware will support at most eight sections.
Thus, eight bits should be enough for long term needs.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Co-developed-by: Serge Ayoun <serge.ayoun@intel.com>
Signed-off-by: Serge Ayoun <serge.ayoun@intel.com>
Co-developed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-6-jarkko@kernel.org
ENCLS is the userspace instruction which wraps virtually all
unprivileged SGX functionality for managing enclaves. It is essentially
the ioctl() of instructions with each function implementing different
SGX-related functionality.
Add macros to wrap the ENCLS functionality. There are two main groups,
one for functions which do not return error codes and a “ret_” set for
those that do.
ENCLS functions are documented in Intel SDM section 36.6.
Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-3-jarkko@kernel.org
Define the SGX architectural data structures used by various SGX
functions. This is not an exhaustive representation of all SGX data
structures but only those needed by the kernel.
The goal is to sequester hardware structures in "sgx/arch.h" and keep
them separate from kernel-internal or uapi structures.
The data structures are described in Intel SDM section 37.6.
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Jethro Beekman <jethro@fortanix.com>
Link: https://lkml.kernel.org/r/20201112220135.165028-2-jarkko@kernel.org
Currently, scan_microcode() leverages microcode_matches() to check
if the microcode matches the CPU by comparing the family and model.
However, the processor stepping and flags of the microcode signature
should also be considered when saving a microcode patch for early
update.
Use find_matching_signature() in scan_microcode() and get rid of the
now-unused microcode_matches() which is a good cleanup in itself.
Complete the verification of the patch being saved for early loading in
save_microcode_patch() directly. This needs to be done there too because
save_mc_for_early() will call save_microcode_patch() too.
The second reason why this needs to be done is because the loader still
tries to support, at least hypothetically, mixed-steppings systems and
thus adds all patches to the cache that belong to the same CPU model
albeit with different steppings.
For example:
microcode: CPU: sig=0x906ec, pf=0x2, rev=0xd6
microcode: mc_saved[0]: sig=0x906e9, pf=0x2a, rev=0xd6, total size=0x19400, date = 2020-04-23
microcode: mc_saved[1]: sig=0x906ea, pf=0x22, rev=0xd6, total size=0x19000, date = 2020-04-27
microcode: mc_saved[2]: sig=0x906eb, pf=0x2, rev=0xd6, total size=0x19400, date = 2020-04-23
microcode: mc_saved[3]: sig=0x906ec, pf=0x22, rev=0xd6, total size=0x19000, date = 2020-04-27
microcode: mc_saved[4]: sig=0x906ed, pf=0x22, rev=0xd6, total size=0x19400, date = 2020-04-23
The patch which is being saved for early loading, however, can only be
the one which fits the CPU this runs on so do the signature verification
before saving.
[ bp: Do signature verification in save_microcode_patch()
and rewrite commit message. ]
Fixes: ec400ddeff ("x86/microcode_intel_early.c: Early update ucode on Intel's CPU")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: stable@vger.kernel.org
Link: https://bugzilla.kernel.org/show_bug.cgi?id=208535
Link: https://lkml.kernel.org/r/20201113015923.13960-1-yu.c.chen@intel.com
Booting as a guest under KVM results in error messages about
unchecked MSR access:
unchecked MSR access error: RDMSR from 0x17f at rIP: 0xffffffff84483f16 (mce_intel_feature_init+0x156/0x270)
because KVM doesn't provide emulation for random model specific
registers.
Switch to using rdmsrl_safe()/wrmsrl_safe() to avoid the message.
Fixes: 68299a42f8 ("x86/mce: Enable additional error logging on certain Intel CPUs")
Reported-by: Qian Cai <cai@redhat.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201111003954.GA11878@agluck-desk2.amr.corp.intel.com
Currently, accessing /proc/cpuinfo sends IPIs to idle CPUs in order to
learn their clock frequency. Which is a bit strange, given that waking
them from idle likely significantly changes their clock frequency.
This commit therefore avoids sending /proc/cpuinfo-induced IPIs to
idle CPUs.
[ paulmck: Also check for idle in arch_freq_prepare_all(). ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
The aperfmperf_snapshot_cpu() function is invoked upon access to
/proc/cpuinfo, and it does do an early exit if the specified CPU has
recently done a snapshot. Unfortunately, the indication that a snapshot
has been completed is set in an IPI handler, and the execution of this
handler can be delayed by any number of unfortunate events. This means
that a system that starts a number of applications, each of which
parses /proc/cpuinfo, can suffer from an smp_call_function_single()
storm, especially given that each access to /proc/cpuinfo invokes
smp_call_function_single() for all CPUs. Please note that this is not
theoretical speculation. Note also that one CPU's pending IPI serves
all requests, so there is no point in ever having more than one IPI
pending to a given CPU.
This commit therefore suppresses duplicate IPIs to a given CPU via a
new ->scfpending field in the aperfmperf_sample structure. This field
is set to the value one if an IPI is pending to the corresponding CPU
and to zero otherwise.
The aperfmperf_snapshot_cpu() function uses atomic_xchg() to set this
field to the value one and sample the old value. If this function's
"wait" parameter is zero, smp_call_function_single() is called only if
the old value of the ->scfpending field was zero. The IPI handler uses
atomic_set_release() to set this new field to zero just before returning,
so that the prior stores into the aperfmperf_sample structure are seen
by future requests that get to the atomic_xchg(). Future requests that
pass the elapsed-time check are ordered by the fact that on x86 loads act
as acquire loads, just as was the case prior to this change. The return
value is based off of the age of the prior snapshot, just as before.
Reported-by: Dave Jones <davej@codemonkey.org.uk>
[ paulmck: Allow /proc/cpuinfo to take advantage of arch_freq_get_on_cpu(). ]
[ paulmck: Add comment on memory barrier. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
On AMD CPUs which have the feature X86_FEATURE_AMD_STIBP_ALWAYS_ON,
STIBP is set to on and
spectre_v2_user_stibp == SPECTRE_V2_USER_STRICT_PREFERRED
At the same time, IBPB can be set to conditional.
However, this leads to the case where it's impossible to turn on IBPB
for a process because in the PR_SPEC_DISABLE case in ib_prctl_set() the
spectre_v2_user_stibp == SPECTRE_V2_USER_STRICT_PREFERRED
condition leads to a return before the task flag is set. Similarly,
ib_prctl_get() will return PR_SPEC_DISABLE even though IBPB is set to
conditional.
More generally, the following cases are possible:
1. STIBP = conditional && IBPB = on for spectre_v2_user=seccomp,ibpb
2. STIBP = on && IBPB = conditional for AMD CPUs with
X86_FEATURE_AMD_STIBP_ALWAYS_ON
The first case functions correctly today, but only because
spectre_v2_user_ibpb isn't updated to reflect the IBPB mode.
At a high level, this change does one thing. If either STIBP or IBPB
is set to conditional, allow the prctl to change the task flag.
Also, reflect that capability when querying the state. This isn't
perfect since it doesn't take into account if only STIBP or IBPB is
unconditionally on. But it allows the conditional feature to work as
expected, without affecting the unconditional one.
[ bp: Massage commit message and comment; space out statements for
better readability. ]
Fixes: 21998a3515 ("x86/speculation: Avoid force-disabling IBPB based on STIBP and enhanced IBRS.")
Signed-off-by: Anand K Mistry <amistry@google.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lkml.kernel.org/r/20201105163246.v2.1.Ifd7243cd3e2c2206a893ad0a5b9a4f19549e22c6@changeid
Lockdep state handling on NMI enter and exit is nothing specific to X86. It's
not any different on other architectures. Also the extra state type is not
necessary, irqentry_state_t can carry the necessary information as well.
Move it to common code and extend irqentry_state_t to carry lockdep state.
[ Ira: Make exit_rcu and lockdep a union as they are mutually exclusive
between the IRQ and NMI exceptions, and add kernel documentation for
struct irqentry_state_t ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20201102205320.1458656-7-ira.weiny@intel.com
When a Linux VM runs on Hyper-V, if the VM has CPUs with >255 APIC IDs,
the CPUs can't be the destination of IOAPIC interrupts, because the
IOAPIC RTE's Dest Field has only 8 bits. Currently the hackery driver
drivers/iommu/hyperv-iommu.c is used to ensure IOAPIC interrupts are
only routed to CPUs that don't have >255 APIC IDs. However, there is
an issue with kdump, because the kdump kernel can run on any CPU, and
hence IOAPIC interrupts can't work if the kdump kernel run on a CPU
with a >255 APIC ID.
The kdump issue can be fixed by the Extended Dest ID, which is introduced
recently by David Woodhouse (for IOAPIC, see the field virt_destid_8_14 in
struct IO_APIC_route_entry). Of course, the Extended Dest ID needs the
support of the underlying hypervisor. The latest Hyper-V has added the
support recently: with this commit, on such a Hyper-V host, Linux VM
does not use hyperv-iommu.c because hyperv_prepare_irq_remapping()
returns -ENODEV; instead, Linux kernel's generic support of Extended Dest
ID from David is used, meaning that Linux VM is able to support up to
32K CPUs, and IOAPIC interrupts can be routed to all the CPUs.
On an old Hyper-V host that doesn't support the Extended Dest ID, nothing
changes with this commit: Linux VM is still able to bring up the CPUs with
> 255 APIC IDs with the help of hyperv-iommu.c, but IOAPIC interrupts still
can not go to such CPUs, and the kdump kernel still can not work properly
on such CPUs.
[ tglx: Updated comment as suggested by David ]
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lore.kernel.org/r/20201103011136.59108-1-decui@microsoft.com
The Xeon versions of Sandy Bridge, Ivy Bridge and Haswell support an
optional additional error logging mode which is enabled by an MSR.
Previously, this mode was enabled from the mcelog(8) tool via /dev/cpu,
but userspace should not be poking at MSRs. So move the enabling into
the kernel.
[ bp: Correct the explanation why this is done. ]
Suggested-by: Boris Petkov <bp@alien8.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201030190807.GA13884@agluck-desk2.amr.corp.intel.com
Intel Memory Bandwidth Monitoring (MBM) counters may report system
memory bandwidth incorrectly on some Intel processors. The errata SKX99
for Skylake server, BDF102 for Broadwell server, and the correction
factor table are documented in Documentation/x86/resctrl.rst.
Intel MBM counters track metrics according to the assigned Resource
Monitor ID (RMID) for that logical core. The IA32_QM_CTR register
(MSR 0xC8E) used to report these metrics, may report incorrect system
bandwidth for certain RMID values.
Due to the errata, system memory bandwidth may not match what is
reported.
To work around the errata, correct MBM total and local readings using a
correction factor table. If rmid > rmid threshold, MBM total and local
values should be multiplied by the correction factor.
[ bp: Mark mbm_cf_table[] __initdata. ]
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lkml.kernel.org/r/20201014004927.1839452-3-fenghua.yu@intel.com
A previous commit changed the notification mode from true/false to an
int, allowing notify-no, notify-yes, or signal-notify. This was
backwards compatible in the sense that any existing true/false user
would translate to either 0 (on notification sent) or 1, the latter
which mapped to TWA_RESUME. TWA_SIGNAL was assigned a value of 2.
Clean this up properly, and define a proper enum for the notification
mode. Now we have:
- TWA_NONE. This is 0, same as before the original change, meaning no
notification requested.
- TWA_RESUME. This is 1, same as before the original change, meaning
that we use TIF_NOTIFY_RESUME.
- TWA_SIGNAL. This uses TIF_SIGPENDING/JOBCTL_TASK_WORK for the
notification.
Clean up all the callers, switching their 0/1/false/true to using the
appropriate TWA_* mode for notifications.
Fixes: e91b481623 ("task_work: teach task_work_add() to do signal_wake_up()")
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Pull another Hyper-V update from Wei Liu:
"One patch from Michael to get VMbus interrupt from ACPI DSDT"
* tag 'hyperv-next-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
Drivers: hv: vmbus: Add parsing of VMbus interrupt in ACPI DSDT
On ARM64, Hyper-V now specifies the interrupt to be used by VMbus
in the ACPI DSDT. This information is not used on x86 because the
interrupt vector must be hardcoded. But update the generic
VMbus driver to do the parsing and pass the information to the
architecture specific code that sets up the Linux IRQ. Update
consumers of the interrupt to get it from an architecture specific
function.
Signed-off-by: Michael Kelley <mikelley@microsoft.com>
Link: https://lore.kernel.org/r/1597434304-40631-1-git-send-email-mikelley@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Pull x86 SEV-ES support from Borislav Petkov:
"SEV-ES enhances the current guest memory encryption support called SEV
by also encrypting the guest register state, making the registers
inaccessible to the hypervisor by en-/decrypting them on world
switches. Thus, it adds additional protection to Linux guests against
exfiltration, control flow and rollback attacks.
With SEV-ES, the guest is in full control of what registers the
hypervisor can access. This is provided by a guest-host exchange
mechanism based on a new exception vector called VMM Communication
Exception (#VC), a new instruction called VMGEXIT and a shared
Guest-Host Communication Block which is a decrypted page shared
between the guest and the hypervisor.
Intercepts to the hypervisor become #VC exceptions in an SEV-ES guest
so in order for that exception mechanism to work, the early x86 init
code needed to be made able to handle exceptions, which, in itself,
brings a bunch of very nice cleanups and improvements to the early
boot code like an early page fault handler, allowing for on-demand
building of the identity mapping. With that, !KASLR configurations do
not use the EFI page table anymore but switch to a kernel-controlled
one.
The main part of this series adds the support for that new exchange
mechanism. The goal has been to keep this as much as possibly separate
from the core x86 code by concentrating the machinery in two
SEV-ES-specific files:
arch/x86/kernel/sev-es-shared.c
arch/x86/kernel/sev-es.c
Other interaction with core x86 code has been kept at minimum and
behind static keys to minimize the performance impact on !SEV-ES
setups.
Work by Joerg Roedel and Thomas Lendacky and others"
* tag 'x86_seves_for_v5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (73 commits)
x86/sev-es: Use GHCB accessor for setting the MMIO scratch buffer
x86/sev-es: Check required CPU features for SEV-ES
x86/efi: Add GHCB mappings when SEV-ES is active
x86/sev-es: Handle NMI State
x86/sev-es: Support CPU offline/online
x86/head/64: Don't call verify_cpu() on starting APs
x86/smpboot: Load TSS and getcpu GDT entry before loading IDT
x86/realmode: Setup AP jump table
x86/realmode: Add SEV-ES specific trampoline entry point
x86/vmware: Add VMware-specific handling for VMMCALL under SEV-ES
x86/kvm: Add KVM-specific VMMCALL handling under SEV-ES
x86/paravirt: Allow hypervisor-specific VMMCALL handling under SEV-ES
x86/sev-es: Handle #DB Events
x86/sev-es: Handle #AC Events
x86/sev-es: Handle VMMCALL Events
x86/sev-es: Handle MWAIT/MWAITX Events
x86/sev-es: Handle MONITOR/MONITORX Events
x86/sev-es: Handle INVD Events
x86/sev-es: Handle RDPMC Events
x86/sev-es: Handle RDTSC(P) Events
...
Pull x86 asm updates from Borislav Petkov:
"Two asm wrapper fixes:
- Use XORL instead of XORQ to avoid a REX prefix and save some bytes
in the .fixup section, by Uros Bizjak.
- Replace __force_order dummy variable with a memory clobber to fix
LLVM requiring a definition for former and to prevent memory
accesses from still being cached/reordered, by Arvind Sankar"
* tag 'x86_asm_for_v5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/asm: Replace __force_order with a memory clobber
x86/uaccess: Use XORL %0,%0 in __get_user_asm()
Pull x86 Hyper-V update from Ingo Molnar:
"A single commit harmonizing the x86 and ARM64 Hyper-V constants
namespace"
* tag 'x86-hyperv-2020-10-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/hyperv: Remove aliases with X64 in their name
Pull x86 paravirt cleanup from Ingo Molnar:
"Clean up the paravirt code after the removal of 32-bit Xen PV support"
* tag 'x86-paravirt-2020-10-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/paravirt: Avoid needless paravirt step clearing page table entries
x86/paravirt: Remove set_pte_at() pv-op
x86/entry/32: Simplify CONFIG_XEN_PV build dependency
x86/paravirt: Use CONFIG_PARAVIRT_XXL instead of CONFIG_PARAVIRT
x86/paravirt: Clean up paravirt macros
x86/paravirt: Remove 32-bit support from CONFIG_PARAVIRT_XXL
