On Sapphire Rapids and related platforms, the DSA and IAA devices have an
erratum that causes direct access (for example, by using the ENQCMD or
MOVDIR64 instructions) from untrusted applications to be a security problem.
To solve this, add a flag to the PCI device enumeration and device structures
to indicate the presence/absence of this security exposure. In the mmap()
method of the device, this flag is then used to enforce that the user
has the CAP_SYS_RAWIO capability.
In a future patch, a write() based method will be added that allows untrusted
applications submit work to the accelerator, where the kernel can do
sanity checking on the user input to ensure secure operation of the accelerator.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
drain_workqueue() cannot be called safely in a spinlocked context due to
possible task rescheduling. In the multi-task scenario, calling
queue_work() while drain_workqueue() will lead to a Call Trace as
pushing a work on a draining workqueue is not permitted in spinlocked
context.
Call Trace:
<TASK>
? __warn+0x7d/0x140
? __queue_work+0x2b2/0x440
? report_bug+0x1f8/0x200
? handle_bug+0x3c/0x70
? exc_invalid_op+0x18/0x70
? asm_exc_invalid_op+0x1a/0x20
? __queue_work+0x2b2/0x440
queue_work_on+0x28/0x30
idxd_misc_thread+0x303/0x5a0 [idxd]
? __schedule+0x369/0xb40
? __pfx_irq_thread_fn+0x10/0x10
? irq_thread+0xbc/0x1b0
irq_thread_fn+0x21/0x70
irq_thread+0x102/0x1b0
? preempt_count_add+0x74/0xa0
? __pfx_irq_thread_dtor+0x10/0x10
? __pfx_irq_thread+0x10/0x10
kthread+0x103/0x140
? __pfx_kthread+0x10/0x10
ret_from_fork+0x31/0x50
? __pfx_kthread+0x10/0x10
ret_from_fork_asm+0x1b/0x30
</TASK>
The current implementation uses a spinlock to protect event log workqueue
and will lead to the Call Trace due to potential task rescheduling.
To address the locking issue, convert the spinlock to mutex, allowing
the drain_workqueue() to be called in a safe mutex-locked context.
This change ensures proper synchronization when accessing the event log
workqueue, preventing potential Call Trace and improving the overall
robustness of the code.
Fixes: c40bd7d973 ("dmaengine: idxd: process user page faults for completion record")
Signed-off-by: Rex Zhang <rex.zhang@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Lijun Pan <lijun.pan@intel.com>
Link: https://lore.kernel.org/r/20240404223949.2885604-1-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add a load_device_defaults() function pointer to struct
idxd_driver_data, which if defined, will be called when an idxd device
is probed and will allow the idxd device to be configured with default
values.
The load_device_defaults() function is passed an idxd device to work
with to set specific device attributes.
Also add a load_device_defaults() implementation IAA devices; future
patches would add default functions for other device types such as
DSA.
The way idxd device probing works, if the device configuration is
valid at that point e.g. at least one workqueue and engine is properly
configured then the device will be enabled and ready to go.
The IAA implementation, idxd_load_iaa_device_defaults(), configures a
single workqueue (wq0) for each device with the following default
values:
mode "dedicated"
threshold 0
size Total WQ Size from WQCAP
priority 10
type IDXD_WQT_KERNEL
group 0
name "iaa_crypto"
driver_name "crypto"
Note that this now adds another configuration step for any users that
want to configure their own devices/workqueus with something different
in that they'll first need to disable (in the case of IAA) wq0 and the
device itself before they can set their own attributes and re-enable,
since they've been already been auto-enabled. Note also that in order
for the new configuration to be applied to the deflate-iaa crypto
algorithm the iaa_crypto module needs to unregister the old version,
which is accomplished by removing the iaa_crypto module, and
re-registering it with the new configuration by reinserting the
iaa_crypto module.
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Kernel workqueues were disabled due to flawed use of kernel VA and SVA
API. Now that we have the support for attaching PASID to the device's
default domain and the ability to reserve global PASIDs from SVA APIs,
we can re-enable the kernel work queues and use them under DMA API.
We also use non-privileged access for in-kernel DMA to be consistent
with the IOMMU settings. Consequently, interrupt for user privilege is
enabled for work completion IRQs.
Link: https://lore.kernel.org/linux-iommu/20210511194726.GP1002214@nvidia.com/
Tested-by: Tony Zhu <tony.zhu@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230802212427.1497170-9-jacob.jun.pan@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
New support:
- Apple admac t8112 device support
- StarFive JH7110 DMA controller
Updates:
- Big pile of idxd updates to support IAA 2.0 device capabilities, DSA
2.0 Event Log and completion record faulting features and new DSA
operations
- at_xdmac supend & resume updates and driver code cleanup
- k3-udma supend & resume support
- k3-psil thread support for J784s4
-----BEGIN PGP SIGNATURE-----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=Vqrw
-----END PGP SIGNATURE-----
Merge tag 'dmaengine-6.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine
Pull dmaengine updates from Vinod Koul:
"New support:
- Apple admac t8112 device support
- StarFive JH7110 DMA controller
Updates:
- Big pile of idxd updates to support IAA 2.0 device capabilities,
DSA 2.0 Event Log and completion record faulting features and
new DSA operations
- at_xdmac supend & resume updates and driver code cleanup
- k3-udma supend & resume support
- k3-psil thread support for J784s4"
* tag 'dmaengine-6.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (57 commits)
dmaengine: idxd: add per wq PRS disable
dmaengine: idxd: add pid to exported sysfs attribute for opened file
dmaengine: idxd: expose fault counters to sysfs
dmaengine: idxd: add a device to represent the file opened
dmaengine: idxd: add per file user counters for completion record faults
dmaengine: idxd: process batch descriptor completion record faults
dmaengine: idxd: add descs_completed field for completion record
dmaengine: idxd: process user page faults for completion record
dmaengine: idxd: add idxd_copy_cr() to copy user completion record during page fault handling
dmaengine: idxd: create kmem cache for event log fault items
dmaengine: idxd: add per DSA wq workqueue for processing cr faults
dmanegine: idxd: add debugfs for event log dump
dmaengine: idxd: add interrupt handling for event log
dmaengine: idxd: setup event log configuration
dmaengine: idxd: add event log size sysfs attribute
dmaengine: idxd: make misc interrupt one shot
dt-bindings: dma: snps,dw-axi-dmac: constrain the items of resets for JH7110 dma
dt-bindings: dma: Drop unneeded quotes
dmaengine: at_xdmac: align declaration of ret with the rest of variables
dmaengine: at_xdmac: add a warning message regarding for unpaused channels
...
The iommu subsystem requires IOMMU_DEV_FEAT_IOPF must be enabled before
and disabled after IOMMU_DEV_FEAT_SVA, if device's I/O page faults rely
on the IOMMU. Add explicit IOMMU_DEV_FEAT_IOPF enabling/disabling in this
driver.
At present, missing IOPF enabling/disabling doesn't cause any real issue,
because the IOMMU driver places the IOPF enabling/disabling in the path
of SVA feature handling. But this may change.
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230324120234.313643-2-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Add event log processing for faulting of user batch descriptor completion
record.
When encountering an event log entry for a page fault on a completion
record, the driver is expected to do the following:
1. If the "first error in batch" bit in event log entry error info is
set, discard any previously recorded errors associated with the
"batch identifier".
2. Fix the page fault according to the fault address in the event log. If
successful, write the completion record to the fault address in user space.
3. If an error is encountered while writing the completion record and it is
associated to a descriptor in the batch, the driver associates the error
with the batch identifier of the event log entry and tracks it until the
event log entry for the corresponding batch desc is encountered.
While processing an event log entry for a batch descriptor with error
indicating that one or more descs in the batch had event log entries,
the driver will do the following before writing the batch completion
record:
1. If the status field of the completion record is 0x1, the driver will
change it to error code 0x5 (one or more operations in batch completed
with status not successful) and changes the result field to 1.
2. If the status is error code 0x6 (page fault on batch descriptor list
address), change the result field to 1.
3. If status is any other value, the completion record is not changed.
4. Clear the recorded error in preparation for next batch with same batch
identifier.
The result field is for user software to determine whether to set the
"Batch Error" flag bit in the descriptor for continuation of partial
batch descriptor completion. See DSA spec 2.0 for additional information.
If no error has been recorded for the batch, the batch completion record is
written to user space as is.
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-12-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
DSA supports page fault handling through PRS. However, the DMA engine
that's processing the descriptor is blocked until the PRS response is
received. Other workqueues sharing the engine are also blocked.
Page fault handing by the driver with PRS disabled can be used to
mitigate the stalling.
With PRS disabled while ATS remain enabled, DSA handles page faults on
a completion record by reporting an event in the event log. In this
instance, the descriptor is completed and the event log contains the
completion record address and the contents of the completion record. Add
support to the event log handling code to fault in the completion record
and copy the content of the completion record to user memory.
A bitmap is introduced to keep track of discarded event log entries. When
the user process initiates ->release() of the char device, it no longer is
interested in any remaining event log entries tied to the relevant wq and
PASID. The driver will mark the event log entry index in the bitmap. Upon
encountering the entries during processing, the event log handler will just
clear the bitmap bit and skip the entry rather than attempt to process the
event log entry.
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-10-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Define idxd_copy_cr() to copy completion record to fault address in
user address that is found by work queue (wq) and PASID.
It will be used to write the user's completion record that the hardware
device is not able to write due to user completion record page fault.
An xarray is added to associate the PASID and mm with the
struct idxd_user_context so mm can be found by PASID and wq.
It is called when handling the completion record fault in a kernel thread
context. Switch to the mm using kthread_use_vm() and copy the
completion record to the mm via copy_to_user(). Once the copy is
completed, switch back to the current mm using kthread_unuse_mm().
Suggested-by: Christoph Hellwig <hch@infradead.org>
Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Suggested-by: Tony Luck <tony.luck@intel.com>
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-9-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add a kmem cache per device for allocating event log fault context. The
context allows an event log entry to be copied and passed to a software
workqueue to be processed. Due to each device can have different sized
event log entry depending on device type, it's not possible to have a
global kmem cache.
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-8-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add debugfs entry to dump the content of the event log for debugging. The
function will dump all non-zero entries in the event log. It will note
which entries are processed and which entries are still pending processing
at the time of the dump. The entries may not always be in chronological
order due to the log is a circular buffer.
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-6-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add setup of event log feature for supported device. Event log addresses
error reporting that was lacking in gen 1 DSA devices where a second error
event does not get reported when a first event is pending software
handling. The event log allows a circular buffer that the device can push
error events to. It is up to the user to create a large enough event log
ring in order to capture the expected events. The evl size can be set in
the device sysfs attribute. By default 64 entries are supported as minimal
when event log is enabled.
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-4-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add support for changing of the event log size. Event log is a
feature added to DSA 2.0 hardware to improve error reporting.
It supersedes the SWERROR register on DSA 1.0 hardware and hope
to prevent loss of reported errors.
The error log size determines how many error entries supported for
the device. It can be configured by the user via sysfs attribute.
Tested-by: Tony Zhu <tony.zhu@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230407203143.2189681-3-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add IAA (IAX) capability mask sysfs attribute to expose to applications.
The mask provides application knowledge of what capabilities this IAA
device supports. This mask is available for IAA 2.0 device or later.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20230303213732.3357494-3-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
INVALID_IOASID and IOMMU_PASID_INVALID are duplicated. Rename
INVALID_IOASID and consolidate since we are moving away from IOASID
infrastructure.
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230322200803.869130-7-jacob.jun.pan@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Currently default read buffers that is allowed in a group is 0.
grpcfg will be configured to max read buffers that IDXD can support if
the group's allowed read buffers value is 0. But 0 is an invalid
read buffers value and user may get confused when seeing the invalid
initial value 0 through sysfs interface.
To show only valid allowed read buffers value and eliminate confusion,
directly initialize the allowed read buffers to IDXD's max read buffers.
User still can change the value through sysfs interface.
Suggested-by: Ramesh Thomas <ramesh.thomas@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20230127192855.966929-1-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
On DSA/IAX 1.0, TC-A and TC-B in GRPCFG are set as 1 to have best
performance and cannot be changed through sysfs knobs unless override
option is given.
The same values should be set on DSA 2.0 as well.
Fixes: ea7c8f598c ("dmaengine: idxd: restore traffic class defaults after wq reset")
Fixes: ade8a86b51 ("dmaengine: idxd: Set defaults for GRPCFG traffic class")
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20221209172141.562648-1-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Resolve conflicts in drivers/vfio/vfio_main.c by using the iommfd version.
The rc fix was done a different way when iommufd patches reworked this
code.
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
>From Intel IAA spec [1], Intel IAA does not support batch processing.
Two batch related default values for IAA are incorrect in current code:
(1) The max batch size of device is set during device initialization,
that indicates batch is supported. It should be always 0 on IAA.
(2) The max batch size of work queue is set to WQ_DEFAULT_MAX_BATCH (32)
as the default value regardless of Intel DSA or IAA device during
work queue setup and cleanup. It should be always 0 on IAA.
Fix the issues by setting the max batch size of device and max batch
size of work queue to 0 on IAA device, that means batch is not
supported.
[1]: https://cdrdv2.intel.com/v1/dl/getContent/721858
Fixes: 23084545db ("dmaengine: idxd: set max_xfer and max_batch for RO device")
Fixes: 92452a72eb ("dmaengine: idxd: set defaults for wq configs")
Fixes: bfe1d56091 ("dmaengine: idxd: Init and probe for Intel data accelerators")
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220930201528.18621-2-xiaochen.shen@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The current kernel DMA with PASID support is based on the SVA with a flag
SVM_FLAG_SUPERVISOR_MODE. The IOMMU driver binds the kernel memory address
space to a PASID of the device. The device driver programs the device with
kernel virtual address (KVA) for DMA access. There have been security and
functional issues with this approach:
- The lack of IOTLB synchronization upon kernel page table updates.
(vmalloc, module/BPF loading, CONFIG_DEBUG_PAGEALLOC etc.)
- Other than slight more protection, using kernel virtual address (KVA)
has little advantage over physical address. There are also no use
cases yet where DMA engines need kernel virtual addresses for in-kernel
DMA.
This removes SVM_FLAG_SUPERVISOR_MODE support from the IOMMU interface.
The device drivers are suggested to handle kernel DMA with PASID through
the kernel DMA APIs.
The drvdata parameter in iommu_sva_bind_device() and all callbacks is not
needed anymore. Cleanup them as well.
Link: https://lore.kernel.org/linux-iommu/20210511194726.GP1002214@nvidia.com/
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Tested-by: Tony Zhu <tony.zhu@intel.com>
Link: https://lore.kernel.org/r/20221031005917.45690-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
DSA 2.0 add the capability of configuring DMA ops on a per workqueue basis.
This means that certain ops can be disabled by the system administrator for
certain wq. By default, all ops are available. A bitmap is used to store
the ops due to total op size of 256 bits and it is more convenient to use a
range list to specify which bits are enabled.
One of the usage to support this is for VM migration between different
iteration of devices. The newer ops are disabled in order to allow guest to
migrate to a host that only support older ops. Another usage is to
restrict the WQ to certain operations for QoS of performance.
A sysfs of ops_config attribute is added per wq. It is only usable when the
ops_config bit is set under WQ_CAP register. This means that this attribute
will return -EOPNOTSUPP on DSA 1.x devices. The expected input is a range
list for the bits per operation the WQ supports.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220917161222.2835172-4-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
To make input and output consistent and prepping for the per WQ operation
configuration support, change the output of opcap display to match the
input that is expected by bitmap_parse() helper function. The output will
be a bitmap with field width as the number of bits using the %*pb format
specifier for printk() family.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220917161222.2835172-3-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Now that idxd_wq_disable_cleanup() sets the workqueue state to
IDXD_WQ_DISABLED, use a bitmap to track which workqueues have been
enabled. This will then be used to determine which workqueues
should be re-enabled when attempting a software reset to recover
from a device halt state.
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20220928154856.623545-3-jsnitsel@redhat.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The idxd driver always gated the pasid enabling under a single knob and
this assumption is incorrect. The pasid used for kernel operation can be
independently toggled and has no dependency on the user pasid (and vice
versa). Split the two so they are independent "enabled" flags.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/165231431746.986466.5666862038354800551.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
DSA spec v1.2 has changed the term of "bandwidth tokens" to "read buffers"
in order to make the concept clearer. Deprecate bandwidth token
naming in the driver and convert to read buffers in order to match with
the spec and reduce confusion when reading the spec.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163951338932.2988321.6162640806935567317.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Change the driver where WQ interrupt is requested only when wq is being
enabled. This new scheme set things up so that request_threaded_irq() is
only called when a kernel wq type is being enabled. This also sets up for
future interrupt request where different interrupt handler such as wq
occupancy interrupt can be setup instead of the wq completion interrupt.
Not calling request_irq() until the WQ actually needs an irq also prevents
wasting of CPU irq vectors on x86 systems, which is a limited resource.
idxd_flush_pending_descs() is moved to device.c since descriptor flushing
is now part of wq disable rather than shutdown().
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163942149487.2412839.6691222855803875848.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The descriptor flushing for shutdown is not holding the irq_entry list
lock. If there's ongoing interrupt completion handling, this can corrupt
the list. Add locking to protect list walking. Also refactor the code so
it's more compact.
Fixes: 8f47d1a5e5 ("dmaengine: idxd: connect idxd to dmaengine subsystem")
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163942148935.2412839.18282664745572777280.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
With irq_entry already being associated with the wq in a 1:1 relationship,
embed the irq_entry in the idxd_wq struct and remove back pointers for
idxe_wq and idxd_device. In the process of this work, clean up the interrupt
handle assignment so that there's no decision to be made during submit
call on where interrupt handle value comes from. Set the interrupt handle
during irq request initialization time.
irq_entry 0 is designated as special and is tied to the device itself.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163942148362.2412839.12055447853311267866.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add a sysfs knob to allow tuning of retries for the kernel ENQCMDS
descriptor submission. While on host, it is not as likely that ENQCMDS
return busy during normal operations due to the driver controlling the
number of descriptors allocated for submission. However, when the driver is
operating as a guest driver, the chance of retry goes up significantly due
to sharing a wq with multiple VMs. A default value is provided with the
system admin being able to tune the value on a per WQ basis.
Suggested-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163820629464.2702134.7577370098568297574.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
"Interrupt handle revoked" is an event that happens when the driver is
running on a guest kernel and the VM is migrated to a new machine.
The device will trigger an interrupt that signals to the guest driver
that the interrupt handles need to be replaced.
The misc irq thread function calls a helper function to handle the
event. The function uses the WQ percpu_ref to quiesce the kernel
submissions. It then replaces the interrupt handles by requesting
interrupt handle command for each I/O MSIX vector. Once the handle is
updated, the driver will unblock the submission path to allow new
submissions.
The submitter will attempt to acquire a percpu_ref before submission. When
the request fails, it will wait on the wq_resurrect 'completion'.
The driver does anticipate the possibility of descriptors being submitted
before the WQ percpu_ref is killed. If a descriptor has already been
submitted, it will return with incorrect interrupt handle status. The
descriptor will be re-submitted with the new interrupt handle on the
completion path. For descriptors with incorrect interrupt handles,
completion interrupt won't be triggered.
At the completion of the interrupt handle refresh, the handling function
will call idxd_int_handle_refresh_drain() to issue drain descriptors to
each of the wq with associated interrupt handle. The drain descriptor will have
interrupt request set but without completion record. This will ensure all
descriptors with incorrect interrupt completion handle get drained and
a completion interrupt is triggered for the guest driver to process them.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Co-Developed-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163528420189.3925689.18212568593220415551.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Attach int_handle to irq_entry. This removes the separate management of int
handles and reduces the confusion of interating through int handles that is
off by 1 count.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163528417065.3925689.11505755433684476288.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Refactor the completion function to allow skipping of descriptor freeing on
the submission failure path. This completely removes descriptor freeing
from the submit failure path and leave the responsibility to the caller.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/163528416222.3925689.12859769271667814762.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Set GRPCFG traffic class to value of 1 for best performance on current
generation of accelerators. Also add override option to allow experimentation.
Sysfs knobs are disabled for DSA/IAX gen1 devices.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162681373005.1968485.3761065664382799202.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The original architecture of /sys/bus/dsa invented a scheme whereby
a single entry in the list of bus drivers, /sys/bus/drivers/dsa,
handled all device types and internally routed them to different
different drivers. Those internal drivers were invisible to
userspace.
With the idxd driver transitioned to a proper bus device-driver model,
the legacy behavior needs to be preserved due to it being exposed to
user space via sysfs. Create a compat driver to provide the legacy
behavior for /sys/bus/dsa/drivers/dsa. This should satisfy user
tool accel-config v3.2 or ealier where this behavior is expected.
If the distro has a newer accel-config then the legacy mode does
not need to be enabled.
When the compat driver binds the device (i.e. dsa0) to the dsa driver,
it will be bound to the new idxd_drv. The wq device (i.e. wq0.0) will
be bound to either the dmaengine_drv or the user_drv. The dsa_drv
becomes a routing mechansim for the new drivers. It will not support
additional external drivers that are implemented later.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637468705.744545.4399080971745974435.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
In preparation for dsa_drv compat support to be built-in, move the bus
code to its own compilation unit. A follow-on patch adds the compat
implementation. Recall that the compat implementation allows for the
deprecated / omnibus dsa_drv binding scheme rather than the idiomatic
organization of a full fledged bus driver per driver type.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637468142.744545.2811632736881720857.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The original architecture of /sys/bus/dsa invented a scheme whereby a
single entry in the list of bus drivers, /sys/bus/drivers/dsa, handled
all device types and internally routed them to different drivers.
Those internal drivers were invisible to userspace. Now, as
/sys/bus/dsa wants to grow support for alternate drivers for a given
device, for example vfio-mdev instead of kernel-internal-dmaengine, a
proper bus device-driver model is needed. The first step in that process
is separating the existing omnibus/implicit "dsa" driver into proper
individual drivers registered on /sys/bus/dsa. Establish the
idxd_user_drv driver that controls the enabling and disabling of the
wq and also register and unregister a char device to allow user space
to mmap the descriptor submission portal.
The cdev related bits are moved to the cdev driver probe/remove and out of
the drv_enabe/disable_wq() calls. These bits are exclusive to the cdev
operation and not part of the generic enable/disable of the wq device.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637467578.744545.10203997610072341376.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The original architecture of /sys/bus/dsa invented a scheme whereby a
single entry in the list of bus drivers, /sys/bus/drivers/dsa, handled
all device types and internally routed them to different drivers.
Those internal drivers were invisible to userspace. Now, as
/sys/bus/dsa wants to grow support for alternate drivers for a given
device, for example vfio-mdev instead of kernel-internal-dmaengine, a
proper bus device-driver model is needed. The first step in that process
is separating the existing omnibus/implicit "dsa" driver into proper
individual drivers registered on /sys/bus/dsa. Establish the
idxd_dmaengine_drv driver that controls the enabling and disabling of the
wq and also register and unregister the dma channel.
idxd_wq_alloc_resources() and idxd_wq_free_resources() also get moved to
the dmaengine driver. The resources (dma descriptors allocation and setup)
are only used by the dmaengine driver and should only happen when it loads.
The char dev driver (cdev) related bits are left in the __drv_enable_wq()
and __drv_disable_wq() calls to be moved when we split out the char dev
driver just like how the dmaengine driver is split out.
WQ autoload support is not expected currently. With the amount of
configuration needed for the device, the wq is always expected to
be enabled by a tool (or via sysfs) rather than auto enabled at driver
load.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637467033.744545.12330636655625405394.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The original architecture of /sys/bus/dsa invented a scheme whereby a
single entry in the list of bus drivers, /sys/bus/drivers/dsa, handled
all device types and internally routed them to different drivers.
Those internal drivers were invisible to userspace. Now, as
/sys/bus/dsa wants to grow support for alternate drivers for a given
device, for example vfio-mdev instead of kernel-internal-dmaengine, a
proper bus device-driver model is needed. The first step in that process
is separating the existing omnibus/implicit "dsa" driver into proper
individual drivers registered on /sys/bus/dsa. Establish the idxd_drv
driver that control the enabling and disabling of the accelerator device.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637466439.744545.15210886092627144577.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add an array of support device types to the idxd_device_driver
definition in order to enable simple matching of device type to a
given driver. The deprecated / omnibus dsa_drv driver specifies
IDXD_DEV_NONE as its only role is to service legacy userspace (old
accel-config) directed bind requests and route them to them the proper
driver. It need not attach to a device when the bus is autoprobed. The
accel-config tooling is being updated to drop its dependency on this
deprecated bind scheme.
Reviewed-by: Dan Willliams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637465882.744545.17456174666211577867.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The IDXD_DEV_CONF_READY state flag is no longer needed. The current
implementation uses this flag to stop the device from doing
configuration until the pci driver probe has completed. With the
driver architecture going towards multiple sub-driver attached to
the dsa_bus, this is no longer feasible. The sub-drivers will be
allowed to probe and return with failure when they are not ready
to complete the probe rather than using a state flag to gate the
probing.
There is no expectation that the devices auto-attach to a driver.
Userspace configuration is expected to setup the device before
enabling.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637460633.744545.8902095097471365420.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add a 'struct idxd_dev' that wraps the 'struct device' for idxd conf_dev
that registers with the dsa bus. This is introduced in order to deal with
multiple different types of 'devices' that are registered on the dsa_bus
when the compat driver needs to route them to the correct driver to attach.
The bind() call now can determine the type of device and then do the
appropriate driver matching.
Reviewed-by Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/162637460065.744545.584492831446090984.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
