IPC use cases with RVD set can't convey proper agent handles.
Runtime discovery is required to properly route the copy in this
case.
Change-Id: I4c97e132fb4b6ac1040de1cb17fe5a3e36d6be48
This is consistent with KFD and has significantly better latency.
KFD is taking this as the definition of the SystemClockCounter.
Change-Id: I4c1b3bc58c738206265c55ebefd41356c013bfe5
The start iterator becomes invalid after it is removed from
std::map prefetch_map_. This was causing a segfault when the iterator is
incremented afterwards.
Signed-off-by: David Yat Sin <david.yatsin@amd.com>
Change-Id: I4b0b763d2cb4ee99c0b8571c2c526b834e74077a
Prior solution used a single global lock to protect the memory tracking structures.
This change protects the memory tracking structure with a shared mutex (rw lock) in
shared (r) mode for memory allocations and frees so that long duration processes,
calling to kfd, can be done in parallel. Operations which must modify the memory map
take the mutex in exclusive mode (w) and must not call to the thunk while holding
the mutex.
The fragment allocator now requires separate protection and is protected with a
mutex at the device level. Protecting at the device level, rather than pool,
allows retention of the current recursive design and allows calling Trim from
withing Allocate. This could be made finer (pool level locks) but would
require backing out of Allocate entirely to call Trim. Trim and any retried
Allocation must be done in isolation (per device) or we may report OOM when
memory is actually available in some pool's fragment cache. So some device
level serialization is required in at least some paths.
Change-Id: I7c1e94d6965ffcc602b12fefdd3a6e97b84b5e00
Early exit if the range is found to be fine grain. Indeterminate
should only apply if the range is neither coarse nor fine.
Change-Id: I54133e14f4e8cfa53e2d612f6112cdcdb5a47dfa
Take in const void* rather than void*. This does not break the
abi or existing code. Existing code would need to cast away any
const which is unnecessary and annoying.
Change-Id: I28787e8fab1b600bf6871ea82835e10a4f475c5b
global_flags reporting allows discovery of an allocation's memory
model (coarse, fine, kernarg). This is critical on gfx90a and
also allows discovery of the memory model of IPC imports.
Change-Id: Icbc3c243ca20e264af5e1931becd2419f762c7ad
Previously ranges were reported as fine if and only if they were
entirely fine. Coarse and mixed ranges were reported as coarse.
For gfx90a it is critical to know if a range is coarse or fine as
fp atomics targeting fine do not function. Range queried reporting
coarse must be able to be trusted so must only report coarse if the
entire region is coarse.
Change-Id: I29c654a2afcd6943961eb2455e3654dfdb1283b5
Certain special signals do not carry their updates via their signal
value. These signals are wrappers around special KFD events, of
which the only current instance informs about VM faults. We either
need to check each signal for this special event type or rely on
the checking done in hsa_amd_signal_wait_any. Since there will always
be a small number of these signals it doesn't make much since to
penalize the performance path with this check. Additionally we know
that the signal indicated by hsa_amd_signal_wait_any is satisfied so
don't need to recheck it's conditions.
Change-Id: I9fc6298300ad543d823ecd28ca8fab4ad26c23ef
Under high async handler load signal retention and event sorting
become bottlenecks. This change processes more handlers in a
single pass to amortize wait_any overheads.
Change-Id: I8b276e102db647e3858e120547aa0c6fca85ab4c
Prevents poorly written tools which throw in tools interface
callbacks from causing ROCr to catch and return a generic error
code.
Change-Id: I2f5bf7104dc7d4ee688eb48423c7ffdb06bd7702
Includes some workarounds and HMM.
Conflicts:
opensrc/hsa-runtime/core/runtime/amd_topology.cpp
opensrc/hsa-runtime/core/util/flag.h
Change-Id: I22976f07964a43dbb228a6231777dbd599112b8d
Sramecc is misreported in kfd 4.0 and prior. To prevent possible
corruption due to d16 instructions, deny use of gfx906 with older
kfds and correct misreport for gfx908. Denial of gfx906 may be
overridden by setting HSA_IGNORE_SRAMECC_MISREPORT=1.
Change-Id: I7d5c3a716fad01c348f8b88cd508cedbf914c989
Memory from the suballocator may be exported via IPC. If this
happens then the allocating process should not reuse that memory
since it would still be connected to the remote process. IPC exported
memory must be released back to the driver.
Change-Id: I2ab0c814f63191f753fc3640cc4140ee144bf07f
All types which could be generated from a fragment need to take this branch.
Taking the branch is correct for all types, it was a performance optimization
only and was missing IPC. Branch removal simplifies updates for any future
fragment use and will allow CQE to report any performance issues that might
require bringing the branch back.
Change-Id: I8041788c422e880b764e144eb1877f5126ba76f3
Thunk may report nullptr for host base if the host does not have
access. Use agent base in this case.
Change-Id: I44883d35a3fff0941b1e3037d16b059591a6c511
IPC create must only be used on whole ROCr allocations.
Fragments were allowing handle creation with offsets.
Change-Id: I1faa96d36bc7a6199bdc2e3ff1b8871d1a36a2fa
This has been the default mode for a while now since we don't
distribute or build the finalizer. Removing the attempt cleans
up debug mode messages that are causing confusion.
Change-Id: I8162c95abd5bbedaa22b90191f7a384a34c388ae
Lock API suceeds but the GPU still faults on the address.
This should be fixed in Thunk and/or KFD as well.
Change-Id: I8b2fbcae61ab181e4fe7f0b64e43a5f0772efb24
Remove agent lookup in time stamp translation for IPC signals. The copy
agent handle is not shared so does not need to be checked for cross
process use. Cross process copy-timestamp read is illegal and continues
to deliver garbage.
Store the copy agent properly when doing CPU-CPU copies.
Change-Id: Ib4008f66ff866922047749dd556c84a32021c1fd
agentOwner from thunk reflects the GPU which holds the device alias.
We need to return a CPU to better reflect that the memory is system memory.
Change-Id: I9233f8779a4bfd471f68dbbbce07ae4528412e18
Adds hsa_amd_register_deallocation_callback and hsa_amd_deregister_deallocation_callback
to notify when HSA memory has been released.
Change-Id: I1f33cee250ca890e5c2e7fddfa4479aa5874651d
CPUClockCounter is not NTP adjusted (CLOCK_MONOTONIC_RAW) so should be
better for measurements. However, it is implemented with syscall while
CLOCK_MONOTONIC is implemented via vDSO. The latency increase becomes
significant when language layers make corresponding clock measurements.
Reverting to CLOCK_MONOTONIC will reduce latency and allow small
duration events to be measured at the cost of incorporating NTP
frequency skew errors. NTP may adjust frequency by 500ppm so limits us
to ~3 decimals in elapsed time.
Change-Id: I920b9f707f47109d80d6c256c475638c03fb8d76
Small times may be given to time conversion if GPU clocks are used to
accumulate elapsed time. Because HSA APIs deal in absolute time this
leads to large conversion offsets of order system uptime. Variation
in relative clock ratio estimation may be amplified in this case,
destroying elapsed time measurements.
This patch fixes the relative clock ratio used for times which predate
the call to hsa_init. This correlates errors in such times allowing
the elapsed time to be correctly computed.
The effective maximum system uptime before elapsed time conversion becomes
inaccurate is ~3.5 months. GPU event timestamps are good for process uptime
of ~3.5 months. These are limited by double's mantissa precision.
Change-Id: I48752ff354920439d91016d6f2b0c8ddfa60b445
Also rename blit_agent to region_gpu and add comments to clarify
its role in deprecated region API support rather than to do blits.
Change-Id: I80b1043db2e1c5d40a58fc801eef70a688ea9169
During registration we must not call any function that depends on registered
data as the lists are not yet complete. This includes signal allocation since
allocating shared GPU mapped memory depends on the list of GPUs.
Change-Id: I94d59e847802c546c2a5a0d9f55fe5ac3fd1d878
Sean Keely
David Yat Sin
Konstantin Zhuravlyov
Ramesh Errabolu