Add support for separate ring-buffer for metadata packets when using
compute queues. Userspace application allocate the metadata ring-buffer
and the queue ring-buffer with a single allocation. The metadata
ring-buffer starts after the queue ring-buffer.
Signed-off-by: David Yat Sin <[email protected]>
Reviewed-by: Philip Yang <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
GFX 1250 allows the debugger to subcribe to LDS out-of-range read/write
memory violations.
Bump IOCTL minor version and flag KFD capabilities for enablement
hint.
Signed-off-by: Jonathan Kim <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
commit 0f0c8a6983db ("drm/amdkfd: Trap handler support for expert
scheduling mode") introduced support for a trap handler when expert
scheduling mode. However userspace needs to know whether or not a trap
handler support is present.
Bump the KFD IOCTL API so that userspace can key off this to decide.
Suggested-by: Stella Laurenzo <[email protected]>
Fixes: 4238888794 ("drm/amdkfd: Trap handler support for expert scheduling mode")
Reviewed-by: Kent Russell <[email protected]>
Signed-off-by: Mario Limonciello <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
This commit implemetns a new ioctl AMDKFD_IOC_CREATE_PROCESS
that creates a new secondary kfd_progress on the FD.
To keep backward compatibility, userspace programs need to invoke
this ioctl explicitly on a FD to create a secondary
kfd_process which replacing its primary kfd_process.
This commit bumps ioctl minor version.
Signed-off-by: Zhu Lingshan <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
rocm-smi will be able to show the events for KFD process
start/end, it is the implementation of this feature.
Signed-off-by: Eric Huang <[email protected]>
Reviewed-by: Kent Russell <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Add support for more per-process flags starting with option to configure
MFMA precision for gfx 9.5
v2: Change flag name to KFD_PROC_FLAG_MFMA_HIGH_PRECISION
Remove unused else condition
v3: Bump the KFD API version
v4: Missed SH_MEM_CONFIG__PRECISION_MODE__SHIFT define. Added it.
Signed-off-by: Harish Kasiviswanathan <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Reviewed-by: Amber Lin <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
If queue size is less than minimum, clamp it to minimum to prevent
underflow when writing queue mqd.
Signed-off-by: David Yat Sin <[email protected]>
Reviewed-by: Jay Cornwall <[email protected]>
Reviewed-by: Harish Kasiviswanathan <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Add new SMI event to report the dropped event count.
When the event kfifo is full, drop count is not zero, or no enough space
left to store the event message, increase drop count.
After reading event out from kfifo, if event was dropped, drop_count is
not zero, generate a dropped event record and reset drop count to zero.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: James Zhu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
If page migration failed, also output migrate end event to match with
migrate start event, with failure error_code added to the end of the
migrate message macro. This will not break uAPI because application uses
old message macro sscanf drop and ignore the error_code.
Output GPU page fault restore end event if migration failed.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: James Zhu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Document how to use SMI system management interface to enable and
receive SVM events. Document SVM event triggers.
Define SVM events message string format macro that could be used by user
mode for sscanf to parse the event. Add it to uAPI header file to make
it obvious that is changing uAPI in future.
No functional changes.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: James Zhu <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Certain GPUs have better copy performance over xGMI on specific
SDMA engines depending on the source and destination GPU.
Allow users to create SDMA queues on these recommended engines.
Close to 2x overall performance has been observed with this
optimization.
Signed-off-by: Jonathan Kim <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Bump the kfd ioctl minor version to delcare the contiguous VRAM
allocation flag support.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
RDMA device with limited scatter-gather ability requires contiguous VRAM
buffer allocation for RDMA peer direct support.
Add a new KFD alloc memory flag and store as bo alloc flag
AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS. When pin this bo to export for RDMA
peerdirect access, this will set TTM_PL_FLAG_CONTIFUOUS flag, and ask
VRAM buddy allocator to get contiguous VRAM.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Due to a CP interrupt bug, bad packet garbage exception codes are raised.
Do a range check so that the debugger and runtime do not receive garbage
codes.
Update the user api to guard exception code type checking as well.
Signed-off-by: Jonathan Kim <[email protected]>
Tested-by: Jesse Zhang <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
This is not strictly a change in the IOCTL API. This version bump is meant
to indicate to user mode the presence of a number of changes and fixes
that enable the management of VA mappings in compute VMs using the GEM_VA
ioctl for DMABufs exported from KFD.
Signed-off-by: Felix Kuehling <[email protected]>
Reviewed-by: Xiaogang Chen<[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
These flags (for GEM and SVM allocations) allocate
memory that allows for system-scope atomic semantics.
On GFX943 these flags cause caches to be avoided on
non-local memory.
On all other ASICs they are identical in functionality to the
equivalent COHERENT flags.
Corresponding Thunk patch is at
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/pull/88
Reviewed-by: David Yat Sin <[email protected]>
Signed-off-by: David Francis <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Bump the minor version to declare debugging capability is now
available.
Signed-off-by: Jonathan Kim <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Introduce the GPU debug operations interface.
For ROCm-GDB to extend the GNU Debugger's ability to inspect the AMD GPU
instruction set, provide the necessary interface to allow the debugger
to HW debug-mode set and query exceptions per HSA queue, process or
device.
The runtime_enable interface coordinates exception handling with the
HSA runtime.
Usage is available in the kern docs at uapi/linux/kfd_ioctl.h.
Signed-off-by: Jonathan Kim <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Recently introduced commit "drm/amdgpu: Set cache coherency
for GC 9.4.3" did not update the settings applicable for svm ranges.
Add the coherence settings for svm ranges for GFX IP 9.4.3.
Reviewed-by: Amber Lin <[email protected]>
Signed-off-by: Rajneesh Bhardwaj <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Exports a DMA buf fd of a given KFD buffer handle. This is intended for
being able to import KFD BOs into GEM contexts to leverage the
amdgpu_bo_va API for more flexible virtual address mappings. It will
also be used for the new upstreamable RDMA solution coming to UCX and
RCCL.
The corresponding user mode change (Thunk API and kfdtest) is here:
https://github.com/fxkamd/ROCT-Thunk-Interface/commits/fxkamd/dmabuf
Signed-off-by: Felix Kuehling <[email protected]>
Acked-by: Christian König <[email protected]>
Reviewed-by: Xiaogang Chen <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Define new system management interface event IDs for migration, GPU
recoverable page fault, user queues eviction, restore and unmap from
GPU events and corresponding event triggers, those will be implemented
in the following patches.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Add a new KFD ioctl to return the largest possible memory size that
can be allocated as a buffer object using
kfd_ioctl_alloc_memory_of_gpu. It attempts to use exactly the same
accept/reject criteria as that function so that allocating a new
buffer object of the size returned by this new ioctl is guaranteed to
succeed, barring races with other allocating tasks.
This IOCTL will be used by libhsakmt:
https://www.mail-archive.com/[email protected]/msg75743.html
Signed-off-by: Daniel Phillips <[email protected]>
Signed-off-by: David Yat Sin <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Export dmabuf handles for GTT BOs so that their contents can be accessed
using SDMA during checkpoint/restore.
v2: Squash in fix from David to set dmabuf handle to invalid for BOs
that cannot be accessed using SDMA during checkpoint/restore.
Signed-off-by: David Yat Sin <[email protected]>
Reviewed-by : Rajneesh Bhardwaj <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
To remove duplicate code, unify event message format and simplify new
event add in the following patches.
Use KFD_SMI_EVENT_MSG_SIZE to define msg size, the same size will be
used in user space to alloc the msg receive buffer.
Signed-off-by: Philip Yang <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Cleanup the kfd code by removing the unused old debugger
implementation.
The address watch was only ever implemented in the upstream
driver for GFXv7 (Kaveri). The user mode tools runtime using
this API was never open-sourced. Work on the old debugger
prototype that used this API has been discontinued years ago.
Only a small piece of resetting wavefronts is kept and
is moved to kfd_device_queue_manager.c.
Signed-off-by: Mukul Joshi <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Checkpoint-Restore in userspace (CRIU) is a powerful tool that can
snapshot a running process and later restore it on same or a remote
machine but expects the processes that have a device file (e.g. GPU)
associated with them, provide necessary driver support to assist CRIU
and its extensible plugin interface. Thus, In order to support the
Checkpoint-Restore of any ROCm process, the AMD Radeon Open Compute
Kernel driver, needs to provide a set of new APIs that provide
necessary VRAM metadata and its contents to a userspace component
(CRIU plugin) that can store it in form of image files.
This introduces some new ioctls which will be used to checkpoint-Restore
any KFD bound user process. KFD only allows ioctl calls from the same
process that opened the KFD file descriptor. Since these ioctls are
expected to be called from a KFD criu plugin which has elevated ptrace
attached privileges and CAP_CHECKPOINT_RESTORE capabilities attached with
the file descriptors so modify KFD to allow such calls.
(API redesigned by David Yat Sin)
Suggested-by: Felix Kuehling <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: David Yat Sin <[email protected]>
Signed-off-by: Rajneesh Bhardwaj <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Currently the SVM get_attr call allows querying, which flags are set
in the entire address range. Add the opposite query, which flags are
clear in the entire address range. Both queries can be combined in a
single get_attr call, which allows answering questions such as, "is
this address range coherent, non-coherent, or a mix of both"?
Proposed userspace for UAPI:
https://github.com/RadeonOpenCompute/ROCR-Runtime/tree/memory_model_queries
Signed-off-by: Felix Kuehling <[email protected]>
Reviewed-by: Philip Yand <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Xnack retries are used for page fault recovery. Some AMD chip
families support continuously retry while page table entries are invalid.
The driver must handle the page fault interrupt and fill in a valid entry
for the GPU to continue.
This ioctl allows to enable/disable XNACK retries per KFD process.
Signed-off-by: Alex Sierra <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Add svm (shared virtual memory) ioctl data structure and API definition.
The svm ioctl API is designed to be extensible in the future. All
operations are provided by a single IOCTL to preserve ioctl number
space. The arguments structure ends with a variable size array of
attributes that can be used to set or get one or multiple attributes.
Signed-off-by: Philip Yang <[email protected]>
Signed-off-by: Alex Sierra <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Replace spaces with Tabs to fix indentation in kfd_smi_event
enum.
Signed-off-by: Mukul Joshi <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Add support for reporting GPU reset events through SMI. KFD
would report both pre and post GPU reset events.
Signed-off-by: Mukul Joshi <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Event bitmask is a 64-bit mask with only 1 bit set. Sending this
event bitmask in KFD SMI event message is both wasteful of memory
and potentially limiting to only 64 events. Instead send event
index in SMI event message.
Please note this change does not break the ABI for the two event
types defined so far. The new index is identical to the mask used
before.
Signed-off-by: Mukul Joshi <[email protected]>
Suggested-by: Felix Kuehling <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Add support for reporting thermal throttling events through SMI.
Also, add a counter to count the number of throttling interrupts
observed and report the count in the SMI event message.
Signed-off-by: Mukul Joshi <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
When the compute is malfunctioning or performance drops, the system admin
will use SMI (System Management Interface) tool to monitor/diagnostic what
went wrong. This patch provides an event watch interface for the user
space to register devices and subscribe events they are interested. After
registered, the user can use annoymous file descriptor's poll function
with wait-time specified and wait for events to happen. Once an event
happens, the user can use read() to retrieve information related to the
event.
VM fault event is done in this patch.
v2: - remove UNREGISTER and add event ENABLE/DISABLE
- correct kfifo usage
- move event message API to kfd_ioctl.h
v3: send the event msg in text than in binary
v4: support multiple clients
v5: move events enablement from ioctl to fd write
v6: sparse fix
Signed-off-by: Amber Lin <[email protected]>
Reviewed-by: Felix Kuehling <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
Jonathan Kim