The page table pages corresponding to broken down large pages are zapped in
FIFO order, so that the large page can potentially be recovered, if it is
not longer being used for execution. This removes the performance penalty
for walking deeper EPT page tables.
By default, one large page will last about one hour once the guest
reaches a steady state.
Signed-off-by: Junaid Shahid <junaids@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add a function to create a kernel thread associated with a given VM. In
particular, it ensures that the worker thread inherits the priority and
cgroups of the calling thread.
Signed-off-by: Junaid Shahid <junaids@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With some Intel processors, putting the same virtual address in the TLB
as both a 4 KiB and 2 MiB page can confuse the instruction fetch unit
and cause the processor to issue a machine check resulting in a CPU lockup.
Unfortunately when EPT page tables use huge pages, it is possible for a
malicious guest to cause this situation.
Add a knob to mark huge pages as non-executable. When the nx_huge_pages
parameter is enabled (and we are using EPT), all huge pages are marked as
NX. If the guest attempts to execute in one of those pages, the page is
broken down into 4K pages, which are then marked executable.
This is not an issue for shadow paging (except nested EPT), because then
the host is in control of TLB flushes and the problematic situation cannot
happen. With nested EPT, again the nested guest can cause problems shadow
and direct EPT is treated in the same way.
[ tglx: Fixup default to auto and massage wording a bit ]
Originally-by: Junaid Shahid <junaids@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
VMX already does so if the host has SMEP, in order to support the combination of
CR0.WP=1 and CR4.SMEP=1. However, it is perfectly safe to always do so, and in
fact VMX also ends up running with EFER.NXE=1 on old processors that lack the
"load EFER" controls, because it may help avoiding a slow MSR write.
SVM does not have similar code, but it should since recent AMD processors do
support SMEP. So this patch makes the code for the two vendors simpler and
more similar, while fixing an issue with CR0.WP=1 and CR4.SMEP=1 on AMD.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Joerg Roedel <jroedel@suse.de> Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Note that in such a case it is quite likely that KVM will BUG_ON
in __pte_list_remove when the VM is closed. However, there is no
immediate risk of memory corruption in the host so a WARN_ON is
enough and it lets you gather traces for debugging.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After the previous patch, the low bits of the gfn are masked in
both FNAME(fetch) and __direct_map, so we do not need to clear them
in transparent_hugepage_adjust.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
These two functions are basically doing the same thing through
kvm_mmu_get_page, link_shadow_page and mmu_set_spte; yet, for historical
reasons, their code looks very different. This patch tries to take the
best of each and make them very similar, so that it is easy to understand
changes that apply to both of them.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Release the page at the call-site where it was originally acquired.
This makes the exit code cleaner for most call sites, since they
do not need to duplicate code between success and the failure
label.
Signed-off-by: Junaid Shahid <junaids@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It doesn't seem as if there is any particular need for kvm_lock to be a
spinlock, so convert the lock to a mutex so that sleepable functions (in
particular cond_resched()) can be called while holding it.
Signed-off-by: Junaid Shahid <junaids@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The largepages debugfs entry is incremented/decremented as shadow
pages are created or destroyed. Clearing it will result in an
underflow, which is harmless to KVM but ugly (and could be
misinterpreted by tools that use debugfs information), so make
this particular statistic read-only.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: kvm-ppc@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A kernel module may need to check the value of the "mitigations=" kernel
command line parameter as part of its setup when the module needs
to perform software mitigations for a CPU flaw.
Uninline and export the helper functions surrounding the cpu_mitigations
enum to allow for their usage from a module.
Lastly, privatize the enum and cpu_mitigations variable since the value of
cpu_mitigations can be checked with the exported helper functions.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some processors may incur a machine check error possibly resulting in an
unrecoverable CPU lockup when an instruction fetch encounters a TLB
multi-hit in the instruction TLB. This can occur when the page size is
changed along with either the physical address or cache type. The relevant
erratum can be found here:
There are other processors affected for which the erratum does not fully
disclose the impact.
This issue affects both bare-metal x86 page tables and EPT.
It can be mitigated by either eliminating the use of large pages or by
using careful TLB invalidations when changing the page size in the page
tables.
Just like Spectre, Meltdown, L1TF and MDS, a new bit has been allocated in
MSR_IA32_ARCH_CAPABILITIES (PSCHANGE_MC_NO) and will be set on CPUs which
are mitigated against this issue.
For new IBRS_ALL CPUs, the Enhanced IBRS check at the beginning of
cpu_bugs_smt_update() causes the function to return early, unintentionally
skipping the MDS and TAA logic.
This is not a problem for MDS, because there appears to be no overlap
between IBRS_ALL and MDS-affected CPUs. So the MDS mitigation would be
disabled and nothing would need to be done in this function anyway.
But for TAA, the TAA_MSG_SMT string will never get printed on Cascade
Lake and newer.
The check is superfluous anyway: when 'spectre_v2_enabled' is
SPECTRE_V2_IBRS_ENHANCED, 'spectre_v2_user' is always
SPECTRE_V2_USER_NONE, and so the 'spectre_v2_user' switch statement
handles it appropriately by doing nothing. So just remove the check.
There is a general consensus that TSX usage is not largely spread while
the history shows there is a non trivial space for side channel attacks
possible. Therefore the tsx is disabled by default even on platforms
that might have a safe implementation of TSX according to the current
knowledge. This is a fair trade off to make.
There are, however, workloads that really do benefit from using TSX and
updating to a newer kernel with TSX disabled might introduce a
noticeable regressions. This would be especially a problem for Linux
distributions which will provide TAA mitigations.
Introduce config options X86_INTEL_TSX_MODE_OFF, X86_INTEL_TSX_MODE_ON
and X86_INTEL_TSX_MODE_AUTO to control the TSX feature. The config
setting can be overridden by the tsx cmdline options.
Add the documenation for TSX Async Abort. Include the description of
the issue, how to check the mitigation state, control the mitigation,
guidance for system administrators.
[ bp: Add proper SPDX tags, touch ups by Josh and me. ]
Co-developed-by: Antonio Gomez Iglesias <antonio.gomez.iglesias@intel.com> Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com> Signed-off-by: Antonio Gomez Iglesias <antonio.gomez.iglesias@intel.com> Signed-off-by: Borislav Petkov <bp@suse.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Mark Gross <mgross@linux.intel.com> Reviewed-by: Tony Luck <tony.luck@intel.com> Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Platforms which are not affected by X86_BUG_TAA may want the TSX feature
enabled. Add "auto" option to the TSX cmdline parameter. When tsx=auto
disable TSX when X86_BUG_TAA is present, otherwise enable TSX.
More details on X86_BUG_TAA can be found here:
https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/tsx_async_abort.html
[ bp: Extend the arg buffer to accommodate "auto\0". ]
Export the IA32_ARCH_CAPABILITIES MSR bit MDS_NO=0 to guests on TSX
Async Abort(TAA) affected hosts that have TSX enabled and updated
microcode. This is required so that the guests don't complain,
"Vulnerable: Clear CPU buffers attempted, no microcode"
when the host has the updated microcode to clear CPU buffers.
Microcode update also adds support for MSR_IA32_TSX_CTRL which is
enumerated by the ARCH_CAP_TSX_CTRL bit in IA32_ARCH_CAPABILITIES MSR.
Guests can't do this check themselves when the ARCH_CAP_TSX_CTRL bit is
not exported to the guests.
In this case export MDS_NO=0 to the guests. When guests have
CPUID.MD_CLEAR=1, they deploy MDS mitigation which also mitigates TAA.
Add the sysfs reporting file for TSX Async Abort. It exposes the
vulnerability and the mitigation state similar to the existing files for
the other hardware vulnerabilities.
Sysfs file path is:
/sys/devices/system/cpu/vulnerabilities/tsx_async_abort
TSX Async Abort (TAA) is a side channel vulnerability to the internal
buffers in some Intel processors similar to Microachitectural Data
Sampling (MDS). In this case, certain loads may speculatively pass
invalid data to dependent operations when an asynchronous abort
condition is pending in a TSX transaction.
This includes loads with no fault or assist condition. Such loads may
speculatively expose stale data from the uarch data structures as in
MDS. Scope of exposure is within the same-thread and cross-thread. This
issue affects all current processors that support TSX, but do not have
ARCH_CAP_TAA_NO (bit 8) set in MSR_IA32_ARCH_CAPABILITIES.
On CPUs which have their IA32_ARCH_CAPABILITIES MSR bit MDS_NO=0,
CPUID.MD_CLEAR=1 and the MDS mitigation is clearing the CPU buffers
using VERW or L1D_FLUSH, there is no additional mitigation needed for
TAA. On affected CPUs with MDS_NO=1 this issue can be mitigated by
disabling the Transactional Synchronization Extensions (TSX) feature.
A new MSR IA32_TSX_CTRL in future and current processors after a
microcode update can be used to control the TSX feature. There are two
bits in that MSR:
* TSX_CTRL_RTM_DISABLE disables the TSX sub-feature Restricted
Transactional Memory (RTM).
* TSX_CTRL_CPUID_CLEAR clears the RTM enumeration in CPUID. The other
TSX sub-feature, Hardware Lock Elision (HLE), is unconditionally
disabled with updated microcode but still enumerated as present by
CPUID(EAX=7).EBX{bit4}.
The second mitigation approach is similar to MDS which is clearing the
affected CPU buffers on return to user space and when entering a guest.
Relevant microcode update is required for the mitigation to work. More
details on this approach can be found here:
The TSX feature can be controlled by the "tsx" command line parameter.
If it is force-enabled then "Clear CPU buffers" (MDS mitigation) is
deployed. The effective mitigation state can be read from sysfs.
Add a kernel cmdline parameter "tsx" to control the Transactional
Synchronization Extensions (TSX) feature. On CPUs that support TSX
control, use "tsx=on|off" to enable or disable TSX. Not specifying this
option is equivalent to "tsx=off". This is because on certain processors
TSX may be used as a part of a speculative side channel attack.
Carve out the TSX controlling functionality into a separate compilation
unit because TSX is a CPU feature while the TSX async abort control
machinery will go to cpu/bugs.c.
[ bp: - Massage, shorten and clear the arg buffer.
- Clarifications of the tsx= possible options - Josh.
- Expand on TSX_CTRL availability - Pawan. ]
Transactional Synchronization Extensions (TSX) may be used on certain
processors as part of a speculative side channel attack. A microcode
update for existing processors that are vulnerable to this attack will
add a new MSR - IA32_TSX_CTRL to allow the system administrator the
option to disable TSX as one of the possible mitigations.
The CPUs which get this new MSR after a microcode upgrade are the ones
which do not set MSR_IA32_ARCH_CAPABILITIES.MDS_NO (bit 5) because those
CPUs have CPUID.MD_CLEAR, i.e., the VERW implementation which clears all
CPU buffers takes care of the TAA case as well.
[ Note that future processors that are not vulnerable will also
support the IA32_TSX_CTRL MSR. ]
Add defines for the new IA32_TSX_CTRL MSR and its bits.
TSX has two sub-features:
1. Restricted Transactional Memory (RTM) is an explicitly-used feature
where new instructions begin and end TSX transactions.
2. Hardware Lock Elision (HLE) is implicitly used when certain kinds of
"old" style locks are used by software.
Bit 7 of the IA32_ARCH_CAPABILITIES indicates the presence of the
IA32_TSX_CTRL MSR.
There are two control bits in IA32_TSX_CTRL MSR:
Bit 0: When set, it disables the Restricted Transactional Memory (RTM)
sub-feature of TSX (will force all transactions to abort on the
XBEGIN instruction).
Bit 1: When set, it disables the enumeration of the RTM and HLE feature
(i.e. it will make CPUID(EAX=7).EBX{bit4} and
CPUID(EAX=7).EBX{bit11} read as 0).
The other TSX sub-feature, Hardware Lock Elision (HLE), is
unconditionally disabled by the new microcode but still enumerated
as present by CPUID(EAX=7).EBX{bit4}, unless disabled by
IA32_TSX_CTRL_MSR[1] - TSX_CTRL_CPUID_CLEAR.
Similar to AMD bits, set the Intel bits from the vendor-independent
feature and bug flags, because KVM_GET_SUPPORTED_CPUID does not care
about the vendor and they should be set on AMD processors as well.
Suggested-by: Jim Mattson <jmattson@google.com> Reviewed-by: Jim Mattson <jmattson@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a jump_whitelist bitmap is reused, it needs to be cleared.
Currently this is done with memset() and the size calculation assumes
bitmaps are made of 32-bit words, not longs. So on 64-bit
architectures, only the first half of the bitmap is cleared.
If some whitelist bits are carried over between successive batches
submitted on the same context, this will presumably allow embedding
the rogue instructions that we're trying to reject.
Use bitmap_zero() instead, which gets the calculation right.
Fixes: f8c08d8faee5 ("drm/i915/cmdparser: Add support for backward jumps") Signed-off-by: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In some circumstances the RC6 context can get corrupted. We can detect
this and take the required action, that is disable RC6 and runtime PM.
The HW recovers from the corrupted state after a system suspend/resume
cycle, so detect the recovery and re-enable RC6 and runtime PM.
v2: rebase (Mika)
v3:
- Move intel_suspend_gt_powersave() to the end of the GEM suspend
sequence.
- Add commit message.
v4:
- Rebased on intel_uncore_forcewake_put(i915->uncore, ...) API
change.
v5:
- Rebased on latest upstream gt_pm refactoring.
Signed-off-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In BXT/APL, device 2 MMIO reads from MIPI controller requires its PLL
to be turned ON. When MIPI PLL is turned off (MIPI Display is not
active or connected), and someone (host or GT engine) tries to read
MIPI registers, it causes hard hang. This is a hardware restriction
or limitation.
Driver by itself doesn't read MIPI registers when MIPI display is off.
But any userspace application can submit unprivileged batch buffer for
execution. In that batch buffer there can be mmio reads. And these
reads are allowed even for unprivileged applications. If these
register reads are for MIPI DSI controller and MIPI display is not
active during that time, then the MMIO read operation causes system
hard hang and only way to recover is hard reboot. A genuine
process/application won't submit batch buffer like this and doesn't
cause any issue. But on a compromised system, a malign userspace
process/app can generate such batch buffer and can trigger system
hard hang (denial of service attack).
The fix is to lower the internal MMIO timeout value to an optimum
value of 950us as recommended by hardware team. If the timeout is
beyond 1ms (which will hit for any value we choose if MMIO READ on a
DSI specific register is performed without PLL ON), it causes the
system hang. But if the timeout value is lower than it will be below
the threshold (even if timeout happens) and system will not get into
a hung state. This will avoid a system hang without losing any
programming or GT interrupts, taking the worst case of lowest CDCLK
frequency and early DC5 abort into account.
Signed-off-by: Uma Shankar <uma.shankar@intel.com> Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some of the gen instruction macros (e.g. MI_DISPLAY_FLIP) have the
length directly encoded in them. Since these are used directly in
the tables, the Length becomes part of the comparison used for
matching during parsing. Thus, if the cmd being parsed has a
different length to that in the table, it is not matched and the
cmd is accepted via the default variable length path.
Fix by masking out everything except the Opcode in the cmd tables
Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To keep things manageable, the pre-gen9 cmdparser does not
attempt to track any form of nested BB_START's. This did not
prevent usermode from using nested starts, or even chained
batches because the cmdparser is not strictly enforced pre gen9.
Instead, the existence of a nested BB_START would cause the batch
to be emitted in insecure mode, and any privileged capabilities
would not be available.
For Gen9, the cmdparser becomes mandatory (for BCS at least), and
so not providing any form of nested BB_START support becomes
overly restrictive. Any such batch will simply not run.
We make heavy use of backward jumps in igt, and it is much easier
to add support for this restricted subset of nested jumps, than to
rewrite the whole of our test suite to avoid them.
Add the required logic to support limited backward jumps, to
instructions that have already been validated by the parser.
Note that it's not sufficient to simply approve any BB_START
that jumps backwards in the buffer because this would allow an
attacker to embed a rogue instruction sequence within the
operand words of a harmless instruction (say LRI) and jump to
that.
We introduce a bit array to track every instr offset successfully
validated, and test the target of BB_START against this. If the
target offset hits, it is re-written to the same offset in the
shadow buffer and the BB_START cmd is allowed.
Note: This patch deliberately ignores checkpatch issues in the
cmdtables, in order to match the style of the surrounding code.
We'll correct the entire file in one go in a later patch.
v2: set dispatch secure late (Mika)
v3: rebase (Mika)
v4: Clear whitelist on each parse
Minor review updates (Chris)
v5: Correct backward jump batching
v6: fix compilation error due to struct eb shuffle (Mika)
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the next patch we will be adding a second valid
termination condition which will require a small
amount of refactoring to share logic with the BB_END
case.
Refactor all error conditions to jump to a dedicated
exit path, with 'break' reserved only for a successful
parse.
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For gen9 we enable cmdparsing on the BCS ring, specifically
to catch inadvertent accesses to sensitive registers
Unlike gen7/hsw, we use the parser only to block certain
registers. We can rely on h/w to block restricted commands,
so the command tables only provide enough info to allow the
parser to delineate each command, and identify commands that
access registers.
Note: This patch deliberately ignores checkpatch issues in
favour of matching the style of the surrounding code. We'll
correct the entire file in one go in a later patch.
v3: rebase (Mika)
v4: Add RING_TIMESTAMP registers to whitelist (Jon)
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In "drm/i915: Add support for mandatory cmdparsing" we introduced the
concept of mandatory parsing. This allows the cmdparser to be invoked
even when user passes batch_len=0 to the execbuf ioctl's.
However, the cmdparser needs to know the extents of the buffer being
scanned. Refactor the code to ensure the cmdparser uses the actual
object size, instead of the incoming length, if user passes 0.
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For Gen7, the original cmdparser motive was to permit limited
use of register read/write instructions in unprivileged BB's.
This worked by copying the user supplied bb to a kmd owned
bb, and running it in secure mode, from the ggtt, only if
the scanner finds no unsafe commands or registers.
For Gen8+ we can't use this same technique because running bb's
from the ggtt also disables access to ppgtt space. But we also
do not actually require 'secure' execution since we are only
trying to reduce the available command/register set. Instead we
will copy the user buffer to a kmd owned read-only bb in ppgtt,
and run in the usual non-secure mode.
Note that ro pages are only supported by ppgtt (not ggtt), but
luckily that's exactly what we need.
Add the required paths to map the shadow buffer to ppgtt ro for Gen8+
v2: IS_GEN7/IS_GEN (Mika)
v3: rebase
v4: rebase
v5: rebase
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The existing cmdparser for gen7 can be bypassed by specifying
batch_len=0 in the execbuf call. This is safe because bypassing
simply reduces the cmd-set available.
In a later patch we will introduce cmdparsing for gen9, as a
security measure, which must be strictly enforced since without
it we are vulnerable to DoS attacks.
Introduce the concept of 'required' cmd parsing that cannot be
bypassed by submitting zero-length bb's.
v2: rebase (Mika)
v2: rebase (Mika)
v3: fix conflict on engine flags (Mika)
Retroactively stop reporting support for secure batches
through the api for gen6+ so that older binaries trigger
the fallback path instead.
Older binaries use secure batches pre gen6 to access resources
that are not available to normal usermode processes. However,
all known userspace explicitly checks for HAS_SECURE_BATCHES
before relying on the secure batch feature.
Since there are no known binaries relying on this for newer gens
we can kill secure batches from gen6, via I915_PARAM_HAS_SECURE_BATCHES.
v2: rebase (Mika)
v3: rebase (Mika)
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Takashi Iwai <tiwai@suse.de> Cc: Tyler Hicks <tyhicks@canonical.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Chris Wilson <chris.p.wilson@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we are using the cmdparser, we will have to copy the batch and so
stall for the relocations. Rather than prolong that stall by adding more
relocation requests, just use CPU relocations and do the stall upfront.
drivers/gpu/drm/i915/i915_cmd_parser.c:808:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:811:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:814:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:808:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:811:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:814:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:808:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:811:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:814:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:808:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:811:23: error: not an lvalue
drivers/gpu/drm/i915/i915_cmd_parser.c:814:23: error: not an lvalue
If we move the shift into each case not only do we kill the warning from
smatch, but we shrink the code slightly:
The command MEDIA_VFE_STATE checks bits at offset +2 dwords. However, it is
possible to have MEDIA_VFE_STATE command with length = 0 + LENGTH_BIAS = 2.
In that case check_cmd will read bits from the following command, or even past
the end of the buffer.
If the offset ends up outside of the command length, reject the command.
The find_reg function was assuming that there is always at least one table in
reg_tables. It is not always true.
In case of VCS or VECS, the reg_tables is NULL and reg_table_count is 0,
implying that no register-accessing commands are allowed. However, the command
tables include commands such as MI_STORE_REGISTER_MEM. When trying to check
such command, the find_reg would dereference NULL pointer.
Now it will just return NULL meaning that the register was not found and the
command will be rejected.
If the user has created a read-only object, they should not be allowed
to circumvent the write protection by using a GGTT mmapping. Deny it.
Also most machines do not support read-only GGTT PTEs, so again we have
to reject attempted writes. Fortunately, this is known a priori, so we
can at least reject in the call to create the mmap (with a sanity check
in the fault handler).
v2: Check the vma->vm_flags during mmap() to allow readonly access.
v3: Remove VM_MAYWRITE to curtail mprotect()
Testcase: igt/gem_userptr_blits/readonly_mmap* Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Jon Bloomfield <jon.bloomfield@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.william.auld@gmail.com> Cc: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com> #v1 Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-4-chris@chris-wilson.co.uk Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hook up the flags to allow read-only ppGTT mappings for gen8+
v2: Include a selftest to check that writes to a readonly PTE are
dropped
v3: Don't duplicate cpu_check() as we can just reuse it, and even worse
don't wholesale copy the theory-of-operation comment from igt_ctx_exec
without changing it to explain the intention behind the new test!
v4: Joonas really likes magic mystery values
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.william.auld@gmail.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-2-chris@chris-wilson.co.uk Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We can set a bit inside the ppGTT PTE to indicate a page is read-only;
writes from the GPU will be discarded. We can use this to protect pages
and in particular support read-only userptr mappings (necessary for
importing PROT_READ vma).
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.william.auld@gmail.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-1-chris@chris-wilson.co.uk Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
iso_buffer should be set to NULL after use and free in the while loop.
In the case of isochronous URB in the while loop, iso_buffer is
allocated and after sending it to server, buffer is deallocated. And
then, if the next URB in the while loop is not a isochronous pipe,
iso_buffer still holds the previously deallocated buffer address and
kfree tries to free wrong buffer address.
Fixes: ea44d190764b ("usbip: Implement SG support to vhci-hcd and stub driver") Reported-by: kbuild test robot <lkp@intel.com> Reported-by: Julia Lawall <julia.lawall@lip6.fr> Signed-off-by: Suwan Kim <suwan.kim027@gmail.com> Reviewed-by: Julia Lawall <julia.lawall@lip6.fr> Acked-by: Shuah Khan <skhan@linuxfoundation.org> Link: https://lore.kernel.org/r/20191022093017.8027-1-suwan.kim027@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
cgroup writeback tries to refresh the associated wb immediately if the
current wb is dead. This is to avoid keeping issuing IOs on the stale
wb after memcg - blkcg association has changed (ie. when blkcg got
disabled / enabled higher up in the hierarchy).
Unfortunately, the logic gets triggered spuriously on inodes which are
associated with dead cgroups. When the logic is triggered on dead
cgroups, the attempt fails only after doing quite a bit of work
allocating and initializing a new wb.
While c3aab9a0bd91 ("mm/filemap.c: don't initiate writeback if mapping
has no dirty pages") alleviated the issue significantly as it now only
triggers when the inode has dirty pages. However, the condition can
still be triggered before the inode is switched to a different cgroup
and the logic simply doesn't make sense.
Skip the immediate switching if the associated memcg is dying.
This is a simplified version of the following two patches:
Functions like filemap_write_and_wait_range() should do nothing if inode
has no dirty pages or pages currently under writeback. But they anyway
construct struct writeback_control and this does some atomic operations if
CONFIG_CGROUP_WRITEBACK=y - on fast path it locks inode->i_lock and
updates state of writeback ownership, on slow path might be more work.
Current this path is safely avoided only when inode mapping has no pages.
For example generic_file_read_iter() calls filemap_write_and_wait_range()
at each O_DIRECT read - pretty hot path.
This patch skips starting new writeback if mapping has no dirty tags set.
If writeback is already in progress filemap_write_and_wait_range() will
wait for it.
Link: http://lkml.kernel.org/r/156378816804.1087.8607636317907921438.stgit@buzz Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Tejun Heo <tj@kernel.org> Cc: Jens Axboe <axboe@kernel.dk> Cc: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ECC (memory error detection and correction) mechanism can be
activated or not, controlled by the ECCDIS bit in CAN_MECR. When
disabled, updates on indications and reporting registers are stopped.
So if want to disable ECC completely, had better assert ECCDIS bit, not
just mask the related interrupts.
Fixes: cdce844865be ("can: flexcan: add vf610 support for FlexCAN") Signed-off-by: Joakim Zhang <qiangqing.zhang@nxp.com> Cc: linux-stable <stable@vger.kernel.org> Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
The removal of the LDR initialization in the bigsmp_32 APIC code unearthed
a problem in setup_local_APIC().
The code checks unconditionally for a mismatch of the logical APIC id by
comparing the early APIC id which was initialized in get_smp_config() with
the actual LDR value in the APIC.
Due to the removal of the bogus LDR initialization the check now can
trigger on bigsmp_32 APIC systems emitting a warning for every booting
CPU. This is of course a false positive because the APIC is not using
logical destination mode.
Restrict the check and the possibly resulting fixup to systems which are
actually using the APIC in logical destination mode.
[ tglx: Massaged changelog and added Cc stable ]
Fixes: bae3a8d3308 ("x86/apic: Do not initialize LDR and DFR for bigsmp") Signed-off-by: Jan Beulich <jbeulich@suse.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/666d8f91-b5a8-1afd-7add-821e72a35f03@suse.com Signed-off-by: Sasha Levin <sashal@kernel.org>
At least on the i350 there is an annoying behavior that is maybe also
present on 82580 devices, but was probably not noticed yet as MAS is not
widely used.
If no cable is connected on both fiber/copper ports the media auto sense
code will constantly swap between them as part of the watchdog task and
produce many unnecessary kernel log messages.
The swap code responsible for this behavior (switching to fiber) should
not be executed if the current media type is copper and there is no signal
detected on the fiber port. In this case we can safely wait until the
AUTOSENSE_EN bit is cleared.
Signed-off-by: Manfred Rudigier <manfred.rudigier@omicronenergy.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If rndis_filter_open() fails, we need to remove the rndis device created
in earlier steps, before returning an error code. Otherwise, the retry of
netvsc_attach() from its callers will fail and hang.
Fixes: 7b2ee50c0cd5 ("hv_netvsc: common detach logic") Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
When rmmod hip04_eth.ko, we can get the following warning:
Task track: rmmod(1623)>bash(1591)>login(1581)>init(1)
------------[ cut here ]------------
WARNING: CPU: 0 PID: 1623 at kernel/irq/manage.c:1557 __free_irq+0xa4/0x2ac()
Trying to free already-free IRQ 200
Modules linked in: ping(O) pramdisk(O) cpuinfo(O) rtos_snapshot(O) interrupt_ctrl(O) mtdblock mtd_blkdevrtfs nfs_acl nfs lockd grace sunrpc xt_tcpudp ipt_REJECT iptable_filter ip_tables x_tables nf_reject_ipv
CPU: 0 PID: 1623 Comm: rmmod Tainted: G O 4.4.193 #1
Hardware name: Hisilicon A15
[<c020b408>] (rtos_unwind_backtrace) from [<c0206624>] (show_stack+0x10/0x14)
[<c0206624>] (show_stack) from [<c03f2be4>] (dump_stack+0xa0/0xd8)
[<c03f2be4>] (dump_stack) from [<c021a780>] (warn_slowpath_common+0x84/0xb0)
[<c021a780>] (warn_slowpath_common) from [<c021a7e8>] (warn_slowpath_fmt+0x3c/0x68)
[<c021a7e8>] (warn_slowpath_fmt) from [<c026876c>] (__free_irq+0xa4/0x2ac)
[<c026876c>] (__free_irq) from [<c0268a14>] (free_irq+0x60/0x7c)
[<c0268a14>] (free_irq) from [<c0469e80>] (release_nodes+0x1c4/0x1ec)
[<c0469e80>] (release_nodes) from [<c0466924>] (__device_release_driver+0xa8/0x104)
[<c0466924>] (__device_release_driver) from [<c0466a80>] (driver_detach+0xd0/0xf8)
[<c0466a80>] (driver_detach) from [<c0465e18>] (bus_remove_driver+0x64/0x8c)
[<c0465e18>] (bus_remove_driver) from [<c02935b0>] (SyS_delete_module+0x198/0x1e0)
[<c02935b0>] (SyS_delete_module) from [<c0202ed0>] (__sys_trace_return+0x0/0x10)
---[ end trace bb25d6123d849b44 ]---
Currently "rmmod hip04_eth.ko" call free_irq more than once
as devres_release_all and hip04_remove both call free_irq.
This results in a 'Trying to free already-free IRQ' warning.
To solve the problem free_irq has been moved out of hip04_remove.
Signed-off-by: Jiangfeng Xiao <xiaojiangfeng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the highly unlikely event that we fail to allocate either of the
"/txrx" or "/control" workqueues, we should bail cleanly rather than
blindly march on with NULL queue pointer(s) installed in the
'fjes_adapter' instance.
Endpoints with a maxpacket length of 0 are probably useless. They
can't transfer any data, and it's not at all unlikely that an HCD will
crash or hang when trying to handle an URB for such an endpoint.
Currently the USB core does not check for endpoints having a maxpacket
value of 0. This patch adds a check, printing a warning and skipping
over any endpoints it catches.
Now, the USB spec does not rule out endpoints having maxpacket = 0.
But since they wouldn't have any practical use, there doesn't seem to
be any good reason for us to accept them.
This saves us writing the IBS control MSR twice when disabling the
event.
I searched revision guides for all families since 10h, and did not
find occurrence of erratum #420, nor anything remotely similar:
so we isolate the secondary MSR write to family 10h only.
Also unconditionally update the count mask for IBS Op implementations
that have read & writeable current count (CurCnt) fields in addition
to the MaxCnt field. These bits were reserved on prior
implementations, and therefore shouldn't have negative impact.
Signed-off-by: Kim Phillips <kim.phillips@amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Fixes: c9574fe0bdb9 ("perf/x86-ibs: Implement workaround for IBS erratum #420") Link: https://lkml.kernel.org/r/20191023150955.30292-2-kim.phillips@amd.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The loop that reads all the IBS MSRs into *buf stopped one MSR short of
reading the IbsOpData register, which contains the RipInvalid status bit.
Fix the offset_max assignment so the MSR gets read, so the RIP invalid
evaluation is based on what the IBS h/w output, instead of what was
left in memory.
Signed-off-by: Kim Phillips <kim.phillips@amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Borislav Petkov <bp@alien8.de> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Fixes: d47e8238cd76 ("perf/x86-ibs: Take instruction pointer from ibs sample") Link: https://lkml.kernel.org/r/20191023150955.30292-1-kim.phillips@amd.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
layerscape board sometimes reported some usb call trace, that is due to
kernel sent LPM tokerns automatically when it has no pending transfers
and think that the link is idle enough to enter L1, which procedure will
ask usb register has a recovery,then kernel will compare USBx_GFLADJ and
set GFLADJ_30MHZ, GFLADJ_30MHZ_REG until GFLADJ_30MHZ is equal 0x20, if
the conditions were met then issue occur, but whatever the conditions
whether were met that usb is all need keep GFLADJ_30MHZ of value is 0x20
(xhci spec ask use GFLADJ_30MHZ to adjust any offset from clock source
that generates the clock that drives the SOF counter, 0x20 is default
value of it)That is normal logic, so need remove the call trace.
We meet several NULL pointer issues if configfs_composite_unbind
and composite_setup (or composite_disconnect) are running together.
These issues occur when do the function switch stress test, the
configfs_compsoite_unbind is called from user mode by
echo "" to /sys/../UDC entry, and meanwhile, the setup interrupt
or disconnect interrupt occurs by hardware. The composite_setup
will get the cdev from get_gadget_data, but configfs_composite_unbind
will set gadget data as NULL, so the NULL pointer issue occurs.
This concurrent is hard to reproduce by native kernel, but can be
reproduced by android kernel.
In this commit, we introduce one spinlock belongs to structure
gadget_info since we can't use the same spinlock in usb_composite_dev
due to exclusive running together between composite_setup and
configfs_composite_unbind. And one bit flag 'unbind' to indicate the
code is at unbind routine, this bit is needed due to we release the
lock at during configfs_composite_unbind sometimes, and composite_setup
may be run at that time.
composite_dev_cleanup call from the failure of configfs_composite_bind
frees up the cdev->os_desc_req and cdev->req. If the previous calls of
bind and unbind is successful these will carry stale values.
Consider the below sequence of function calls:
configfs_composite_bind()
composite_dev_prepare()
- Allocate cdev->req, cdev->req->buf
composite_os_desc_req_prepare()
- Allocate cdev->os_desc_req, cdev->os_desc_req->buf
configfs_composite_unbind()
composite_dev_cleanup()
- free the cdev->os_desc_req->buf and cdev->req->buf
Next composition switch
configfs_composite_bind()
- If it fails goto err_comp_cleanup will call the
composite_dev_cleanup() function
composite_dev_cleanup()
- calls kfree up with the stale values of cdev->req->buf and
cdev->os_desc_req from the previous configfs_composite_bind
call. The free call on these stale values leads to double free.
Hence, Fix this issue by setting request and buffer pointer to NULL after
kfree.
Fix interrupt storm generated by endpoints when working in FIFO mode.
The TX_COMPLETE interrupt is used only by control endpoints processing.
Do not enable it for other types of endpoints.
Check memory resource existence before releasing it to avoid NULL
pointer dereference
Signed-off-by: Nikhil Badola <nikhil.badola@freescale.com> Reviewed-by: Ran Wang <ran.wang_1@nxp.com> Reviewed-by: Peter Chen <peter.chen@nxp.com> Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
When a macsec interface is created, it increases a refcnt to a lower
device(real device). when macsec interface is deleted, the refcnt is
decreased in macsec_free_netdev(), which is ->priv_destructor() of
macsec interface.
The problem scenario is this.
When nested macsec interfaces are exiting, the exit routine of the
macsec module makes refcnt leaks.
Test commands:
ip link add dummy0 type dummy
ip link add macsec0 link dummy0 type macsec
ip link add macsec1 link macsec0 type macsec
modprobe -rv macsec
[ 208.629433] unregister_netdevice: waiting for macsec0 to become free. Usage count = 1
Steps of exit routine of macsec module are below.
1. Calls ->dellink() in __rtnl_link_unregister().
2. Checks refcnt and wait refcnt to be 0 if refcnt is not 0 in
netdev_run_todo().
3. Calls ->priv_destruvtor() in netdev_run_todo().
Step2 checks refcnt, but step3 decreases refcnt.
So, step2 waits forever.
This patch makes the macsec module do not hold a refcnt of the lower
device because it already holds a refcnt of the lower device with
netdev_upper_dev_link().
Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: Taehee Yoo <ap420073@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
The IFF_BONDING means bonding master or bonding slave device.
->ndo_add_slave() sets IFF_BONDING flag and ->ndo_del_slave() unsets
IFF_BONDING flag.
bond0<--bond1
Both bond0 and bond1 are bonding device and these should keep having
IFF_BONDING flag until they are removed.
But bond1 would lose IFF_BONDING at ->ndo_del_slave() because that routine
do not check whether the slave device is the bonding type or not.
This patch adds the interface type check routine before removing
IFF_BONDING flag.
Test commands:
ip link add bond0 type bond
ip link add bond1 type bond
ip link set bond1 master bond0
ip link set bond1 nomaster
ip link del bond1 type bond
ip link add bond1 type bond
BUG: KCSAN: data-race in update_defense_level / update_defense_level
read to 0xffffffff861a6260 of 4 bytes by task 3006 on cpu 1:
update_defense_level+0x621/0xb30 net/netfilter/ipvs/ip_vs_ctl.c:177
defense_work_handler+0x3d/0xd0 net/netfilter/ipvs/ip_vs_ctl.c:225
process_one_work+0x3d4/0x890 kernel/workqueue.c:2269
worker_thread+0xa0/0x800 kernel/workqueue.c:2415
kthread+0x1d4/0x200 drivers/block/aoe/aoecmd.c:1253
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:352
write to 0xffffffff861a6260 of 4 bytes by task 7333 on cpu 0:
update_defense_level+0xa62/0xb30 net/netfilter/ipvs/ip_vs_ctl.c:205
defense_work_handler+0x3d/0xd0 net/netfilter/ipvs/ip_vs_ctl.c:225
process_one_work+0x3d4/0x890 kernel/workqueue.c:2269
worker_thread+0xa0/0x800 kernel/workqueue.c:2415
kthread+0x1d4/0x200 drivers/block/aoe/aoecmd.c:1253
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:352
Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 7333 Comm: kworker/0:5 Not tainted 5.4.0-rc3+ #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Workqueue: events defense_work_handler
Indeed, old_secure_tcp is currently a static variable, while it
needs to be a per netns variable.
Fixes: a0840e2e165a ("IPVS: netns, ip_vs_ctl local vars moved to ipvs struct.") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: Simon Horman <horms@verge.net.au> Signed-off-by: Sasha Levin <sashal@kernel.org>
if the IPVS module is removed while the sync daemon is starting, there is
a small gap where try_module_get() might fail getting the refcount inside
ip_vs_use_count_inc(). Then, the refcounts of IPVS module are unbalanced,
and the subsequent call to stop_sync_thread() causes the following splat:
Check the return value of ip_vs_use_count_inc() and let its caller return
proper error. Inside do_ip_vs_set_ctl() the module is already refcounted,
we don't need refcount/derefcount there. Finally, in register_ip_vs_app()
and start_sync_thread(), take the module refcount earlier and ensure it's
released in the error path.
Change since v1:
- better return values in case of failure of ip_vs_use_count_inc(),
thanks to Julian Anastasov
- no need to increase/decrease the module refcount in ip_vs_set_ctl(),
thanks to Julian Anastasov
This patch fixes issue with Gen7 adapter in a blade environment where one
of the ports will not be detected by driver. Firmware expects mailbox 11 to
be set or cleared by driver for newer ISP.
A customer reports that after a devloss, an ADISC failure is logged. It
turns out the ADISC flag is set even the user explicitly set lpfc_use_adisc
= 0.
[Sat Dec 22 22:55:58 2018] lpfc 0000:82:00.0: 2:(0):0203 Devloss timeout on WWPN 50:01:43:80:12:8e:40:20 NPort x05df00 Data: x82000000 x8 xa
[Sat Dec 22 23:08:20 2018] lpfc 0000:82:00.0: 2:(0):2755 ADISC failure DID:05DF00 Status:x9/x70000
[mkp: fixed Hannes' email]
Fixes: 92d7f7b0cde3 ("[SCSI] lpfc: NPIV: add NPIV support on top of SLI-3") Cc: Dick Kennedy <dick.kennedy@broadcom.com> Cc: James Smart <james.smart@broadcom.com> Link: https://lore.kernel.org/r/20191022072112.132268-1-dwagner@suse.de Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: James Smart <james.smart@broadcom.com> Signed-off-by: Daniel Wagner <dwagner@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The function in net core, register_netdevice(), may fail with vport's
destruction callback either invoked or not. After commit 309b66970ee2
("net: openvswitch: do not free vport if register_netdevice() is failed."),
the duty to destroy vport is offloaded from the driver OTOH, which ends
up in the memory leak reported.
It is fixed by releasing vport unless device is registered successfully.
To do that, the callback assignment is defered until device is registered.
Reported-by: syzbot+13210896153522fe1ee5@syzkaller.appspotmail.com Fixes: 309b66970ee2 ("net: openvswitch: do not free vport if register_netdevice() is failed.") Cc: Taehee Yoo <ap420073@gmail.com> Cc: Greg Rose <gvrose8192@gmail.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Cc: Ying Xue <ying.xue@windriver.com> Cc: Andrey Konovalov <andreyknvl@google.com> Signed-off-by: Hillf Danton <hdanton@sina.com> Acked-by: Pravin B Shelar <pshelar@ovn.org>
[sbrivio: this was sent to dev@openvswitch.org and never made its way
to netdev -- resending original patch] Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Greg Rose <gvrose8192@gmail.com> Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The issue is in drivers/infiniband/core/uverbs_std_types_cq.c in the
UVERBS_HANDLER(UVERBS_METHOD_CQ_CREATE) function. We check that:
if (attr.comp_vector >= attrs->ufile->device->num_comp_vectors) {
But we don't check if "attr.comp_vector" is negative. It could
potentially lead to an array underflow. My concern would be where
cq->vector is used in the create_cq() function from the cxgb4 driver.
And really "attr.comp_vector" is appears as a u32 to user space so that's
the right type to use.
Fixes: 9ee79fce3642 ("IB/core: Add completion queue (cq) object actions") Link: https://lore.kernel.org/r/20191011133419.GA22905@mwanda Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In vdma_channel_set_config clear the delay, frame count and master mask
before updating their new values. It avoids programming incorrect state
when input parameters are different from default.
The PCI Tegra controller conversion to a device tree configurable
driver in commit d1523b52bff3 ("PCI: tegra: Move PCIe driver
to drivers/pci/host") implied that code for the driver can be
compiled in for a kernel supporting multiple platforms.
Unfortunately, a blind move of the code did not check that some of the
quirks that were applied in arch/arm (eg enabling Relaxed Ordering on
all PCI devices - since the quirk hook erroneously matches PCI_ANY_ID
for both Vendor-ID and Device-ID) are now applied in all kernels that
compile the PCI Tegra controlled driver, DT and ACPI alike.
This is completely wrong, in that enablement of Relaxed Ordering is only
required by default in Tegra20 platforms as described in the Tegra20
Technical Reference Manual (available at
https://developer.nvidia.com/embedded/downloads#?search=tegra%202 in
Section 34.1, where it is mentioned that Relaxed Ordering bit needs to
be enabled in its root ports to avoid deadlock in hardware) and in the
Tegra30 platforms for the same reasons (unfortunately not documented
in the TRM).
There is no other strict requirement on PCI devices Relaxed Ordering
enablement on any other Tegra platforms or PCI host bridge driver.
Fix this quite upsetting situation by limiting the vendor and device IDs
to which the Relaxed Ordering quirk applies to the root ports in
question, reported above.
There are bugs on vhci with usb 3.0 storage device. In USB, each SG
list entry buffer should be divisible by the bulk max packet size.
But with native SG support, this problem doesn't matter because the
SG buffer is treated as contiguous buffer. But without native SG
support, USB storage driver breaks SG list into several URBs and the
error occurs because of a buffer size of URB that cannot be divided
by the bulk max packet size. The error situation is as follows.
When USB Storage driver requests 31.5 KB data and has SG list which
has 3584 bytes buffer followed by 7 4096 bytes buffer for some
reason. USB Storage driver splits this SG list into several URBs
because VHCI doesn't support SG and sends them separately. So the
first URB buffer size is 3584 bytes. When receiving data from device,
USB 3.0 device sends data packet of 1024 bytes size because the max
packet size of BULK pipe is 1024 bytes. So device sends 4096 bytes.
But the first URB buffer has only 3584 bytes buffer size. So host
controller terminates the transfer even though there is more data to
receive. So, vhci needs to support SG transfer to prevent this error.
In this patch, vhci supports SG regardless of whether the server's
host controller supports SG or not, because stub driver splits SG
list into several URBs if the server's host controller doesn't
support SG.
To support SG, vhci sets URB_DMA_MAP_SG flag in urb->transfer_flags
if URB has SG list and this flag will tell stub driver to use SG
list. After receiving urb from stub driver, vhci clear URB_DMA_MAP_SG
flag to avoid unnecessary DMA unmapping in HCD.
vhci sends each SG list entry to stub driver. Then, stub driver sees
the total length of the buffer and allocates SG table and pages
according to the total buffer length calling sgl_alloc(). After stub
driver receives completed URB, it again sends each SG list entry to
vhci.
If the server's host controller doesn't support SG, stub driver
breaks a single SG request into several URBs and submits them to
the server's host controller. When all the split URBs are completed,
stub driver reassembles the URBs into a single return command and
sends it to vhci.
Moreover, in the situation where vhci supports SG, but stub driver
does not, or vice versa, usbip works normally. Because there is no
protocol modification, there is no problem in communication between
server and client even if the one has a kernel without SG support.
In the case of vhci supports SG and stub driver doesn't, because
vhci sends only the total length of the buffer to stub driver as
it did before the patch applied, stub driver only needs to allocate
the required length of buffers using only kmalloc() regardless of
whether vhci supports SG or not. But stub driver has to allocate
buffer with kmalloc() as much as the total length of SG buffer which
is quite huge when vhci sends SG request, so it has overhead in
buffer allocation in this situation.
If stub driver needs to send data buffer to vhci because of IN pipe,
stub driver also sends only total length of buffer as metadata and
then sends real data as vhci does. Then vhci receive data from stub
driver and store it to the corresponding buffer of SG list entry.
And for the case of stub driver supports SG and vhci doesn't, since
the USB storage driver checks that vhci doesn't support SG and sends
the request to stub driver by splitting the SG list into multiple
URBs, stub driver allocates a buffer for each URB with kmalloc() as
it did before this patch.
* Test environment
Test uses two difference machines and two different kernel version
to make mismatch situation between the client and the server where
vhci supports SG, but stub driver does not, or vice versa. All tests
are conducted in both full SG support that both vhci and stub support
SG and half SG support that is the mismatch situation. Test kernel
version is 5.3-rc6 with commit "usb: add a HCD_DMA flag instead of
guestimating DMA capabilities" to avoid unnecessary DMA mapping and
unmapping.
- Test kernel version
- 5.3-rc6 with SG support
- 5.1.20-200.fc29.x86_64 without SG support
* SG support test
- Test devices
- Super-speed storage device - SanDisk Ultra USB 3.0
- High-speed storage device - SMI corporation USB 2.0 flash drive
- Test description
Test read and write operation of mass storage device that uses the
BULK transfer. In test, the client reads and writes files whose size
is over 1G and it works normally.
* Regression test
- Test devices
- Super-speed device - Logitech Brio webcam
- High-speed device - Logitech C920 HD Pro webcam
- Full-speed device - Logitech bluetooth mouse
- Britz BR-Orion speaker
- Low-speed device - Logitech wired mouse
- Test description
Moving and click test for mouse. To test the webcam, use gnome-cheese.
To test the speaker, play music and video on the client. All works
normally.
* VUDC compatibility test
VUDC also works well with this patch. Tests are done with two USB
gadget created by CONFIGFS USB gadget. Both use the BULK pipe.
1. Serial gadget
2. Mass storage gadget
- Serial gadget test
Serial gadget on the host sends and receives data using cat command
on the /dev/ttyGS<N>. The client uses minicom to communicate with
the serial gadget.
- Mass storage gadget test
After connecting the gadget with vhci, use "dd" to test read and
write operation on the client side.
The patch c6688ef9f297: "usbip: fix stub_rx: harden CMD_SUBMIT path
to handle malicious input" from Dec 7, 2017, leads to the following
static checker warning:
Many kernel drivers contain code that allocates and frees both a
scatterlist and the pages that populate that scatterlist.
Introduce functions in lib/scatterlist.c that perform these tasks
instead of duplicating this functionality in multiple drivers.
Only include these functions in the build if CONFIG_SGL_ALLOC=y
to avoid that the kernel size increases if this functionality is
not used.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.com> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
de53fd7aedb1 ("sched/fair: Fix low cpu usage with high throttling by removing expiration of cpu-local slices")
introduced a few compilation warnings:
kernel/sched/fair.c: In function '__refill_cfs_bandwidth_runtime':
kernel/sched/fair.c:4365:6: warning: variable 'now' set but not used [-Wunused-but-set-variable]
kernel/sched/fair.c: In function 'start_cfs_bandwidth':
kernel/sched/fair.c:4992:6: warning: variable 'overrun' set but not used [-Wunused-but-set-variable]
Also, __refill_cfs_bandwidth_runtime() does no longer update the
expiration time, so fix the comments accordingly.
Signed-off-by: Qian Cai <cai@lca.pw> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Ben Segall <bsegall@google.com> Reviewed-by: Dave Chiluk <chiluk+linux@indeed.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: pauld@redhat.com Fixes: de53fd7aedb1 ("sched/fair: Fix low cpu usage with high throttling by removing expiration of cpu-local slices") Link: https://lkml.kernel.org/r/1566326455-8038-1-git-send-email-cai@lca.pw Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It has been observed, that highly-threaded, non-cpu-bound applications
running under cpu.cfs_quota_us constraints can hit a high percentage of
periods throttled while simultaneously not consuming the allocated
amount of quota. This use case is typical of user-interactive non-cpu
bound applications, such as those running in kubernetes or mesos when
run on multiple cpu cores.
This has been root caused to cpu-local run queue being allocated per cpu
bandwidth slices, and then not fully using that slice within the period.
At which point the slice and quota expires. This expiration of unused
slice results in applications not being able to utilize the quota for
which they are allocated.
The non-expiration of per-cpu slices was recently fixed by
'commit 512ac999d275 ("sched/fair: Fix bandwidth timer clock drift
condition")'. Prior to that it appears that this had been broken since
at least 'commit 51f2176d74ac ("sched/fair: Fix unlocked reads of some
cfs_b->quota/period")' which was introduced in v3.16-rc1 in 2014. That
added the following conditional which resulted in slices never being
expired.
if (cfs_rq->runtime_expires != cfs_b->runtime_expires) {
/* extend local deadline, drift is bounded above by 2 ticks */
cfs_rq->runtime_expires += TICK_NSEC;
Because this was broken for nearly 5 years, and has recently been fixed
and is now being noticed by many users running kubernetes
(https://github.com/kubernetes/kubernetes/issues/67577) it is my opinion
that the mechanisms around expiring runtime should be removed
altogether.
This allows quota already allocated to per-cpu run-queues to live longer
than the period boundary. This allows threads on runqueues that do not
use much CPU to continue to use their remaining slice over a longer
period of time than cpu.cfs_period_us. However, this helps prevent the
above condition of hitting throttling while also not fully utilizing
your cpu quota.
This theoretically allows a machine to use slightly more than its
allotted quota in some periods. This overflow would be bounded by the
remaining quota left on each per-cpu runqueueu. This is typically no
more than min_cfs_rq_runtime=1ms per cpu. For CPU bound tasks this will
change nothing, as they should theoretically fully utilize all of their
quota in each period. For user-interactive tasks as described above this
provides a much better user/application experience as their cpu
utilization will more closely match the amount they requested when they
hit throttling. This means that cpu limits no longer strictly apply per
period for non-cpu bound applications, but that they are still accurate
over longer timeframes.
This greatly improves performance of high-thread-count, non-cpu bound
applications with low cfs_quota_us allocation on high-core-count
machines. In the case of an artificial testcase (10ms/100ms of quota on
80 CPU machine), this commit resulted in almost 30x performance
improvement, while still maintaining correct cpu quota restrictions.
That testcase is available at https://github.com/indeedeng/fibtest.
Fixes: 512ac999d275 ("sched/fair: Fix bandwidth timer clock drift condition") Signed-off-by: Dave Chiluk <chiluk+linux@indeed.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Phil Auld <pauld@redhat.com> Reviewed-by: Ben Segall <bsegall@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: John Hammond <jhammond@indeed.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Kyle Anderson <kwa@yelp.com> Cc: Gabriel Munos <gmunoz@netflix.com> Cc: Peter Oskolkov <posk@posk.io> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Brendan Gregg <bgregg@netflix.com> Link: https://lkml.kernel.org/r/1563900266-19734-2-git-send-email-chiluk+linux@indeed.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The metastability workaround causes Erratic errors [1]
on the HighSpeed USB PHY which can cause upto 2 seconds
delay in enumerating to a USB host while in Gadget mode.
Disable the Run/Stop metastability workaround to avoid this
ill effect. We are aware that this opens up the opportunity
for Run/Stop metastability, however this issue has never been
observed in TI releases so we think that Run/Stop metastability
is a lesser evil than the PHY Erratic errors. So disable it.
A very conservative check for bus activity (to prevent interference
in multimaster setups) prevented the bus recovery methods from being
triggered in the case that SDA or SCL was stuck low.
This defeats the purpose of the recovery mechanism, which was introduced
for exactly this situation (a slave device keeping SDA pulled down).
Also added a check to make sure SDA is low before attempting recovery.
If SDA is not stuck low, recovery will not help, so we can skip it.
Note that bus lockups can persist across reboots. The only other options
are to reset or power cycle the offending slave device, and many i2c
slaves do not even have a reset pin.
If we see that one of the lines is low for the entire timeout duration,
we can actually be sure that there is no other master driving the bus.
It is therefore save for us to attempt a bus recovery.
Signed-off-by: Claudio Foellmi <claudio.foellmi@ergon.ch> Tested-by: Vignesh R <vigneshr@ti.com> Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
[wsa: fixed one return code to -EBUSY] Signed-off-by: Wolfram Sang <wsa@the-dreams.de> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All error handling paths in this function 'goto err' except this one.
If one of the 2 previous memory allocations fails, we should go through
the existing error handling path. Otherwise there is an unbalanced
pm_runtime_enable()/pm_runtime_disable().
Fixes: dd55ff8346a9 ("ASoC: davinci-mcasp: Add set_tdm_slots() support") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com> Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>