commit 2760f5a415 upstream.
Hardware core level testing features require near simultaneous execution
of WRMSR instructions on all threads of a core to initiate a test.
Provide a customized cut down version of stop_machine_cpuslocked() that
just operates on the threads of a single core.
Intel-SIG: commit 2760f5a415 stop_machine: Add stop_core_cpuslocked() for per-core operations
Backport Intel In Field Scan(IFS) single-blob image support.
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20220506225410.1652287-4-tony.luck@intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
[ Aichun Shi: amend commit log ]
Signed-off-by: Aichun Shi <aichun.shi@intel.com>
Long irq latency will affect other latency, such as answer a net
packet. Add a ko to debug long irq latency, account the delay
and show the stack.
Signed-off-by: Jianping Liu <frankjpliu@tencent.com>
(cherry picked from commit 9056a7a86e)
Signed-off-by: Alex Shi <alexsshi@tencent.com>
commit 2a6c6b7d7a upstream.
Current PERF_SAMPLE_WEIGHT sample type is very useful to expresses the
cost of an action represented by the sample. This allows the profiler
to scale the samples to be more informative to the programmer. It could
also help to locate a hotspot, e.g., when profiling by memory latencies,
the expensive load appear higher up in the histograms. But current
PERF_SAMPLE_WEIGHT sample type is solely determined by one factor. This
could be a problem, if users want two or more factors to contribute to
the weight. For example, Golden Cove core PMU can provide both the
instruction latency and the cache Latency information as factors for the
memory profiling.
For current X86 platforms, although meminfo::latency is defined as a
u64, only the lower 32 bits include the valid data in practice (No
memory access could last than 4G cycles). The higher 32 bits can be used
to store new factors.
Add a new sample type, PERF_SAMPLE_WEIGHT_STRUCT, to indicate the new
sample weight structure. It shares the same space as the
PERF_SAMPLE_WEIGHT sample type.
Users can apply either the PERF_SAMPLE_WEIGHT sample type or the
PERF_SAMPLE_WEIGHT_STRUCT sample type to retrieve the sample weight, but
they cannot apply both sample types simultaneously.
Currently, only X86 and PowerPC use the PERF_SAMPLE_WEIGHT sample type.
- For PowerPC, there is nothing changed for the PERF_SAMPLE_WEIGHT
sample type. There is no effect for the new PERF_SAMPLE_WEIGHT_STRUCT
sample type. PowerPC can re-struct the weight field similarly later.
- For X86, the same value will be dumped for the PERF_SAMPLE_WEIGHT
sample type or the PERF_SAMPLE_WEIGHT_STRUCT sample type for now.
The following patches will apply the new factors for the
PERF_SAMPLE_WEIGHT_STRUCT sample type.
The field in the union perf_sample_weight should be shared among
different architectures. A generic name is required, but it's hard to
abstract a name that applies to all architectures. For example, on X86,
the fields are to store all kinds of latency. While on PowerPC, it
stores MMCRA[TECX/TECM], which should not be latency. So a general name
prefix 'var$NUM' is used here.
Intel-SIG: commit 2a6c6b7d7a perf/core: Add PERF_SAMPLE_WEIGHT_STRUCT
Backport for Sapphire Rapids core PMU support.
Note: This backported patch has some deviations from upstream version.
To avoid enum hole in perf_event_sample_format struct, we added
PERF_SAMPLE_{AUX,CGROUP,DATA_PAGE_SIZE,CODE_PAGE_SIZE} to file
include/uapi/linux/perf_event.h, but didn't backport the full patchsets
that introducing these enumeration values. To avoid mishandling of these
sampling formats, we added check to perf_copy_attr() in
kernel/events/core.c, to make sure -EINVAL always being returned for
these lack-of-kernel-support sampling formats.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1611873611-156687-2-git-send-email-kan.liang@linux.intel.com
[ Yunying Sun: amend commit log ]
Signed-off-by: Yunying Sun <yunying.sun@intel.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 2ade0d6093 upstream.
There is plenty of space in the printk_context variable. Reserve one byte
there for the NMI context to be on the safe side.
It should never overflow. The BUG_ON(in_nmi() == NMI_MASK) in nmi_enter()
will trigger much earlier.
Intel-SIG: commit 2ade0d6093 printk: Prepare for nested printk_nmi_enter().
Backport to kernel 5.4 to enhance MCA-R.
Signed-off-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Link: https://lkml.kernel.org/r/20200505134100.681374113@linutronix.de
[ Youquan Song: amend commit log ]
Signed-off-by: Youquan Song <youquan.song@intel.com>
commit opencloudos.
When there's no MSI generated during LM, device dev_msi_list is
actually empty. Add empty check before process the list to prevent
returning an invalid entry pointer, therefore avoiding parsing
invalid memory and eventually eliminating induced error messages.
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 1a5620671a upstream.
With the previous patch, there is an extra watchdog read in each retry.
Now the total number of clocksource reads is increased to 4 per iteration.
In order to avoid increasing the clock skew check overhead, the default
maximum number of retries is reduced from 3 to 2 to maintain the same 12
clocksource reads in the worst case.
Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit c86ff8c55b upstream.
Since commit db3a34e174 ("clocksource: Retry clock read if long delays
detected") and commit 2e27e793e2 ("clocksource: Reduce clocksource-skew
threshold"), it is found that tsc clocksource fallback to hpet can
sometimes happen on both Intel and AMD systems especially when they are
running stressful benchmarking workloads. Of the 23 systems tested with
a v5.14 kernel, 10 of them have switched to hpet clock source during
the test run.
The result of falling back to hpet is a drastic reduction of performance
when running benchmarks. For example, the fio performance tests can
drop up to 70% whereas the iperf3 performance can drop up to 80%.
4 hpet fallbacks happened during bootup. They were:
[ 8.749399] clocksource: timekeeping watchdog on CPU13: hpet read-back delay of 263750ns, attempt 4, marking unstable
[ 12.044610] clocksource: timekeeping watchdog on CPU19: hpet read-back delay of 186166ns, attempt 4, marking unstable
[ 17.336941] clocksource: timekeeping watchdog on CPU28: hpet read-back delay of 182291ns, attempt 4, marking unstable
[ 17.518565] clocksource: timekeeping watchdog on CPU34: hpet read-back delay of 252196ns, attempt 4, marking unstable
Other fallbacks happen when the systems were running stressful
benchmarks. For example:
[ 2685.867873] clocksource: timekeeping watchdog on CPU117: hpet read-back delay of 57269ns, attempt 4, marking unstable
[46215.471228] clocksource: timekeeping watchdog on CPU8: hpet read-back delay of 61460ns, attempt 4, marking unstable
Commit 2e27e793e2 ("clocksource: Reduce clocksource-skew threshold"),
changed the skew margin from 100us to 50us. I think this is too small
and can easily be exceeded when running some stressful workloads on a
thermally stressed system. So it is switched back to 100us.
Even a maximum skew margin of 100us may be too small in for some systems
when booting up especially if those systems are under thermal stress. To
eliminate the case that the large skew is due to the system being too
busy slowing down the reading of both the watchdog and the clocksource,
an extra consecutive read of watchdog clock is being done to check this.
The consecutive watchdog read delay is compared against
WATCHDOG_MAX_SKEW/2. If the delay exceeds the limit, we assume that
the system is just too busy. A warning will be printed to the console
and the clock skew check is skipped for this round.
Fixes: db3a34e174 ("clocksource: Retry clock read if long delays detected")
Fixes: 2e27e793e2 ("clocksource: Reduce clocksource-skew threshold")
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit db3a34e174 upstream.
When the clocksource watchdog marks a clock as unstable, this might be due
to that clock being unstable or it might be due to delays that happen to
occur between the reads of the two clocks. Yes, interrupts are disabled
across those two reads, but there are no shortage of things that can delay
interrupts-disabled regions of code ranging from SMI handlers to vCPU
preemption. It would be good to have some indication as to why the clock
was marked unstable.
Therefore, re-read the watchdog clock on either side of the read from the
clock under test. If the watchdog clock shows an excessive time delta
between its pair of reads, the reads are retried.
The maximum number of retries is specified by a new kernel boot parameter
clocksource.max_cswd_read_retries, which defaults to three, that is, up to
four reads, one initial and up to three retries. If more than one retry
was required, a message is printed on the console (the occasional single
retry is expected behavior, especially in guest OSes). If the maximum
number of retries is exceeded, the clock under test will be marked
unstable. However, the probability of this happening due to various sorts
of delays is quite small. In addition, the reason (clock-read delays) for
the unstable marking will be apparent.
Reported-by: Chris Mason <clm@fb.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Feng Tang <feng.tang@intel.com>
Link: https://lore.kernel.org/r/20210527190124.440372-1-paulmck@kernel.org
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 888f44bd1b53913f7d809369489fd7daae5ae23c Intel-BKC.
Booting multiple VMs causes Invalid notifier called! call trace.
Error message below observed during boot:
[ 638.774647] Invalid notifier called!
[ 638.774669] WARNING: CPU: 319 PID: 46882 at kernel/notifier.c:78 notifier_call_chain+0x79/0xb0
[ 638.870074] RIP: 0010:notifier_call_chain+0x79/0xb0
[ 638.972742] atomic_notifier_call_chain+0x17/0x30
[ 638.989967] ioasid_notify+0x77/0xe0
[ 638.989987] ioasid_alloc+0x19e/0x230
[ 639.003162] ioasid_fops_unl_ioctl+0xb4/0x1b0
This is due to early registration of NULL callback function pointer.
[ 613.023815] ioasid_add_pending_nb: nh ff37c3c3913de380 b call ffffffffc115a770, nr_ioasids 0
[ 613.023883] ioasid_add_pending_nb: nh ff37c3c3913de380 b call 0,
It is unclear who registered the invalid callback, this patch fixes
the issue by rejecting the registration. Dump stack should show the
culprit. It is also possible that nb got cleared somehow outside
ioasid core.
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 86d8cd35c5641a6807506674f72030c4d3ceaa8e Intel-BKC.
Enable a single device to hold both device msi and msi/x interrupts.
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 06fde695ee upstream.
Since commit 5fe71d271d ("irqchip/gic-v3-its: Tag ITS device as shared if
allocating for a proxy device"), some of the devices are wrongly marked as
"shared" by the ITS driver on systems equipped with the ITS(es). The
problem is that the @info->flags may not be initialized anywhere and we end
up looking at random bits on the stack. That's obviously not good.
We can perform the initialization in the IRQ core layer before calling
msi_domain_prepare_irqs(), which is neat enough.
Fixes: 5fe71d271d ("irqchip/gic-v3-its: Tag ITS device as shared if allocating for a proxy device")
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20201218060039.1770-1-yuzenghui@huawei.com
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit f35723a81e0880e3923bc0b3d198bba8d814a3fd Intel-BKC.
Currently, the pci_msix_disable() frees all the allocated resources
associated with a PCIe device when the device is being shut down. With
the introduction of dynamic allocation of MSI-X vectors, there may be
cases where drivers want to free only a particular interrupt, even
when the device is not being shut down.
A new API, pci_free_msix_irq_vector() provides this type of interface.
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit ef818654dbb197aa1e2cd3baaeb0ac1b98b47e26 Intel-BKC.
Introduce a new API pci_add_msix_irq_vector(), which can be called
multiple times by a driver to add a new MSI-X vector to the device
after some number of interrupts have already been allocated using
the existing pci_alloc_irq_vectors API.
If successful, the API returns the device-relative interrupt vector
index (0-based) which can be passed to pci_irq_vector() to retrieve
the Linux IRQ number of that device vector. It should be called only
after the pci_alloc_irq_vectors().
Add a new member msix_alloc_count to keep track of the number of the
MSI-X vectors currently allocated to the device.
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 235988b35dba007134f4ec4d973408b8a5c0c8f5 Intel-BKC.
This is a preparatory patch to enable dynamic allocation of MSI-X
interrupts.
With the addition of dynamic msix, the msi_list of the device is no
longer immutable. To set up new vectors, we need to iterate only through
the newly added entries of this list.
To help with this:
1. A msi_last_list pointer is added to struct device which points to the
last msi_desc's list before a new allocation.
2. New macros are introduced which iterate the msi_list from the 1st
newly added msi_desc of every allocation using (1).
No functional change.
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 84ede51031d8df62a94b4014294e81ff25b502fb Intel-BKC.
Add new helpers to get the Linux IRQ number and device specific index
for given device-relative vector so that the drivers don't need to
allocate their own arrays to keep track of the vectors and hwirq for
the multi vector device MSI case.
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 459da077f38ceffaa524bd504d66739e37314fd1 Intel-BKC.
Introduce a new function pointer in the irq_chip structure(irq_set_auxdata)
which is responsible for updating data which is stored in a shared register
or data storage. For example, the idxd driver uses the auxiliary data API
to enable/set and disable PASID field that is in the IMS entry (introduced
in a later patch) and that data are not typically present in MSI entry.
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit edd679d4da0966b436c15ba1867b995ce582e39b Intel-BKC.
For devices which don't have a standard storage for MSI messages like the
upcoming IMS (Interrupt Message Store) it's required to allocate storage
space before allocating interrupts and after freeing them.
This could be achieved with the existing callbacks, but that would be
awkward because they operate on msi_alloc_info_t which is not uniform
across architectures. Also these callbacks are invoked per interrupt but
the allocation might have bulk requirements depending on the device.
As such devices can operate on different architectures it is simpler to
have separate callbacks which operate on struct device. The resulting
storage information has to be stored in struct msi_desc so the underlying
irq chip implementation can retrieve it for the relevant operations.
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit dbbc93576e upstream.
msi_domain_alloc_irqs() invokes irq_domain_activate_irq(), but
msi_domain_free_irqs() does not enforce deactivation before tearing down
the interrupts.
This happens when PCI/MSI interrupts are set up and never used before being
torn down again, e.g. in error handling pathes. The only place which cleans
that up is the error handling path in msi_domain_alloc_irqs().
Move the cleanup from msi_domain_alloc_irqs() into msi_domain_free_irqs()
to cure that.
Fixes: f3b0946d62 ("genirq/msi: Make sure PCI MSIs are activated early")
Signed-off-by: Bixuan Cui <cuibixuan@huawei.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20210518033117.78104-1-cuibixuan@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 43e9e705dd upstream.
To support MSI irq domains which do not fit at all into the regular MSI
irqdomain scheme, like the XEN MSI interrupt management for PV/HVM/DOM0,
it's necessary to allow to override the alloc/free implementation.
This is a preperatory step to switch X86 away from arch_*_msi_irqs() and
store the irq domain pointer right in struct device.
No functional change for existing MSI irq domain users.
Aside of the evil XEN wrapper this is also useful for special MSI domains
which need to do extra alloc/free work before/after calling the generic
core function. Work like allocating/freeing MSI descriptors, MSI storage
space etc.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20200826112333.526797548@linutronix.de
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 2f4841d50818106b263e60c4fbb6ead6bd770634 Intel-BKC.
Add device specific MSI domain infrastructure for devices which have their
own resource management and interrupt chip. These devices are not related
to PCI and contrary to platform MSI they do not share a common resource and
interrupt chip. They provide their own domain specific resource management
and interrupt chip.
This utilizes the new alloc/free override in a non evil way which avoids
having yet another set of specialized alloc/free functions. Just using
msi_domain_alloc/free_irqs() is sufficient
While initially it was suggested and tried to piggyback device MSI on
platform MSI, the better variant is to reimplement platform MSI on top of
device MSI.
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit d4a4db1e8108a2a78183490484bc9adda867cc0f Intel-BKC.
MSI interrupts have some common flags which should be set not only for
PCI/MSI interrupts.
Move the PCI/MSI flag setting into a common function so it can be reused.
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 2f170814bd upstream.
Move PCI's MSI sysfs code to the irq core so that other busses such as
platform can reuse it.
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210813035628.6844-2-21cnbao@gmail.com
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit a99ffe1a97b9a6151c2f8913fad6da2face81ef8 Intel-BKC.
Until now interrupt chips which support setting affinity are not locking
the associated bus lock for two reasons:
- All chips which support affinity setting do not use buslock because they
just can operated directly on the hardware.
- All chips which use buslock do not support affinity setting because
their interrupt chips are not capable. These chips are usually connected
over a bus like I2C, SPI etc. and have an interrupt output which is
conneted to CPU interrupt of some sort. So there is no way to set the
affinity on the chip itself.
Upcoming hardware which is PCIE based sports a non standard MSI(X) variant
which stores the MSI message in RAM which is associated to e.g. a device
queue. The device manages this RAM and writes have to be issued via command
queues or similar mechanisms which is obviously not possible from interrupt
disabled, raw spinlock held context.
The buslock mechanism of irq chips can be utilized to support that. The
affinity write to the chip writes to shadow state, marks it pending and the
irq chip's irq_bus_sync_unlock() callback handles the command queue and
wait for completion similar to the other chip operations on I2C or SPI
buses.
Change the locking in irq_set_affinity() to bus_lock/unlock to help with
that. There are a few other callers than the proc interface, but none of
them is affected by this change as none of them affects an irq chip with
bus lock support.
Reviewed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Megha Dey <megha.dey@intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 4d80d6ca5d upstream.
Perf modules abuse irq_set_affinity_hint() to set the affinity of system
PMU interrupts just because irq_set_affinity() was not exported.
The fact that irq_set_affinity_hint() actually sets the affinity is a
non-documented side effect and the name is clearly saying it's a hint.
To clean this up, export the real affinity setter.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20210518093117.968251441@linutronix.de
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 72d6878a1c013c048823bf66cce9c00f20d10956 Intel-BKC.
IOASIDs are used to associate DMA requests with virtual address spaces.
They are a system-wide limited resource made available to the userspace
applications. Let it be VMs or user-space device drivers.
This RFC patch introduces a cgroup controller to address the following
problems:
1. Some user applications exhaust all the available IOASIDs thus
depriving others of the same host.
2. System admins need to provision VMs based on their needs for IOASIDs,
e.g. the number of VMs with assigned devices that perform DMA requests
with PASID.
This patch is nowhere near its completion, it merely provides the basic
functionality for resource distribution and cgroup hierarchy
organizational changes.
Since this is part of a greater effort to enable Shared Virtual Address
(SVA) virtualization. We would like to have a direction check and
collect feedback early. For details, please refer to the documentation:
Documentation/admin-guide/cgroup-v1/ioasids.rst
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 95b079d821 upstream.
Fix the type of index from unsigned int to int since find_slots() might
return -1.
Fixes: 26a7e09478 ("swiotlb: refactor swiotlb_tbl_map_single")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Claire Chang <tientzu@chromium.org>
Signed-off-by: Konrad Rzeszutek Wilk <konrad@kernel.org>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 80808d273a upstream.
Split swiotlb_tbl_sync_single into two separate funtions for the to device
and to cpu synchronization.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 2bdba622c3 upstream.
Move the code to find and validate the original buffer address and size
from the callers into swiotlb_bounce. This means a tiny bit of extra
work in the swiotlb_map path, but avoids code duplication and a leads to
a better code structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 2973073a80 upstream.
Now that swiotlb remembers the allocation size there is no need to pass
it back to swiotlb_tbl_unmap_single.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit daf9514fd5 upstream.
The size of the buffer being bounced is not checked if it happens
to be larger than the size of the mapped buffer. Because the size
can be controlled by a device, as it's the case with virtio devices,
this can lead to memory corruption.
This patch saves the remaining buffer memory for each slab and uses
that information for validation in the sync/unmap paths before
swiotlb_bounce is called.
Validating this argument is important under the threat models of
AMD SEV-SNP and Intel TDX, where the HV is considered untrusted.
Signed-off-by: Martin Radev <martin.b.radev@gmail.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 1f221a0d0d upstream.
Respect the min_align_mask in struct device_dma_parameters in swiotlb.
There are two parts to it:
1) for the lower bits of the alignment inside the io tlb slot, just
extent the size of the allocation and leave the start of the slot
empty
2) for the high bits ensure we find a slot that matches the high bits
of the alignment to avoid wasting too much memory
Based on an earlier patch from Jianxiong Gao <jxgao@google.com>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 16fc3cef33 upstream.
swiotlb_tbl_map_single currently nevers sets a tlb_addr that is not
aligned to the tlb bucket size. But we're going to add such a case
soon, for which this adjustment would be bogus.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 26a7e09478 upstream.
Split out a bunch of a self-contained helpers to make the function easier
to follow.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 901c7280ca upstream.
Halil Pasic points out [1] that the full revert of that commit (revert
in bddac7c1e0), and that a partial revert that only reverts the
problematic case, but still keeps some of the cleanups is probably
better. 
And that partial revert [2] had already been verified by Oleksandr
Natalenko to also fix the issue, I had just missed that in the long
discussion.
So let's reinstate the cleanups from commit aa6f8dcbab ("swiotlb:
rework "fix info leak with DMA_FROM_DEVICE""), and effectively only
revert the part that caused problems.
Link: https://lore.kernel.org/all/20220328013731.017ae3e3.pasic@linux.ibm.com/ [1]
Link: https://lore.kernel.org/all/20220324055732.GB12078@lst.de/ [2]
Link: https://lore.kernel.org/all/4386660.LvFx2qVVIh@natalenko.name/ [3]
Suggested-by: Halil Pasic <pasic@linux.ibm.com>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Cc: Christoph Hellwig" <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[OP: backport to 5.4: adjusted context]
Signed-off-by: Ovidiu Panait <ovidiu.panait@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit ddbd89deb7 upstream.
The problem I'm addressing was discovered by the LTP test covering
cve-2018-1000204.
A short description of what happens follows:
1) The test case issues a command code 00 (TEST UNIT READY) via the SG_IO
interface with: dxfer_len == 524288, dxdfer_dir == SG_DXFER_FROM_DEV
and a corresponding dxferp. The peculiar thing about this is that TUR
is not reading from the device.
2) In sg_start_req() the invocation of blk_rq_map_user() effectively
bounces the user-space buffer. As if the device was to transfer into
it. Since commit a45b599ad8 ("scsi: sg: allocate with __GFP_ZERO in
sg_build_indirect()") we make sure this first bounce buffer is
allocated with GFP_ZERO.
3) For the rest of the story we keep ignoring that we have a TUR, so the
device won't touch the buffer we prepare as if the we had a
DMA_FROM_DEVICE type of situation. My setup uses a virtio-scsi device
and the buffer allocated by SG is mapped by the function
virtqueue_add_split() which uses DMA_FROM_DEVICE for the "in" sgs (here
scatter-gather and not scsi generics). This mapping involves bouncing
via the swiotlb (we need swiotlb to do virtio in protected guest like
s390 Secure Execution, or AMD SEV).
4) When the SCSI TUR is done, we first copy back the content of the second
(that is swiotlb) bounce buffer (which most likely contains some
previous IO data), to the first bounce buffer, which contains all
zeros. Then we copy back the content of the first bounce buffer to
the user-space buffer.
5) The test case detects that the buffer, which it zero-initialized,
ain't all zeros and fails.
One can argue that this is an swiotlb problem, because without swiotlb
we leak all zeros, and the swiotlb should be transparent in a sense that
it does not affect the outcome (if all other participants are well
behaved).
Copying the content of the original buffer into the swiotlb buffer is
the only way I can think of to make swiotlb transparent in such
scenarios. So let's do just that if in doubt, but allow the driver
to tell us that the whole mapped buffer is going to be overwritten,
in which case we can preserve the old behavior and avoid the performance
impact of the extra bounce.
Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit ca10d0f8e5 upstream.
Remove a layer of pointless indentation, replace a hard to follow
ternary expression with a plain if/else.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit c32a77fd18 upstream.
Factor out a helper to find the number of slots for a given size.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit c7fbeca757 upstream.
Replace the very genericly named OFFSET macro with a little inline
helper that hardcodes the alignment to the only value ever passed.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit b5d7ccb7aa upstream.
Add a new IO_TLB_SIZE define instead open coding it using
IO_TLB_SHIFT all over.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jianxiong Gao <jxgao@google.com>
Tested-by: Jianxiong Gao <jxgao@google.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit e998879d4f upstream.
For SEV, all DMA to and from guest has to use shared (un-encrypted) pages.
SEV uses SWIOTLB to make this happen without requiring changes to device
drivers. However, depending on the workload being run, the default 64MB
of it might not be enough and it may run out of buffers to use for DMA,
resulting in I/O errors and/or performance degradation for high
I/O workloads.
Adjust the default size of SWIOTLB for SEV guests using a
percentage of the total memory available to guest for the SWIOTLB buffers.
Adds a new sev_setup_arch() function which is invoked from setup_arch()
and it calls into a new swiotlb generic code function swiotlb_adjust_size()
to do the SWIOTLB buffer adjustment.
v5 fixed build errors and warnings as
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Ashish Kalra <ashish.kalra@amd.com>
Co-developed-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 849facea92 upstream.
Use and entirely separate code path for the DMA_ATTR_NO_KERNEL_MAPPING
path. This avoids any confusion about the ret type, and avoids lots of
attr checks and helpers that can be significantly simplified now.
It also ensures that common handling is applied to architetures still
using the arch alloc/free hooks.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 5b138c534f upstream.
This ensures dma_direct_alloc_pages will use the right gfp mask, as
well as keeping the code for that common between the two allocators.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 08a89c2830 upstream.
Check for highmem pages from CMA, just like in the dma_direct_alloc path.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit f959dcd6dd upstream.
When booting the kernel v5.9-rc4 on a VM, the kernel would panic when
printing a warning message in swiotlb_map(). The dev->dma_mask must not
be a NULL pointer when calling the dma mapping layer. A NULL pointer
check can potentially avoid the panic.
Signed-off-by: Thomas Tai <thomas.tai@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 2a047e0662 upstream.
These can only return 0 for failure or the number of entries, so turn
the return value into an unsigned int.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit d03c544192 upstream.
Now that all the .map_sg operations have been converted to returning
proper error codes, drop the code to handle a zero return value,
add a warning if a zero is returned.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
commit 6506932b32 upstream.
The .map_sg() op now expects an error code instead of zero on failure.
The only errno to return is -EINVAL in the case when DMA is not
supported.
Signed-off-by: Martin Oliveira <martin.oliveira@eideticom.com>
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chen Zhuo <sagazchen@tencent.com>
Signed-off-by: Xinghui Li <korantli@tencent.com>
Peter Zijlstra