The kfdtest multi-gpu extension only allows using the first gpu(in topology
order) as default gpu for KFDEvictTest test case. We should allow user
to select any gpu in system as default gpu to run KFDEvictTest test case.
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I8848771d92ae2812fbb96e917454ab3b4a71a2fa
Update KFDMultiProcessTest class to fork process on gpu wise.
Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Change-Id: Ibb12d64b4cbc5f082d737fd8d8a74233b75be13e
Update class KFDBaseComponentTest to have kfdtest run on multi gpu under google test framework.
Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Change-Id: If139c4fc6cf062fb800ff171c69cbf9604f98278
Add HSA_TEST_GPUS_NUM env variable that controls gpu number kfdtest uses.
This env variable will be used by following patches.
Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Change-Id: I800042f93d4e0955040f450e9b5cf09927971323
Extend Google test log functions to extend kfdtest running on multiple gpu.
Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Change-Id: I6d8de92fec606309756423b3a6fa2af3558c0517
Assembler function from LLVM is not multi-thread safe and is ASIC dependent.
To extend kfdtest to test on multiple GPU separate LLVM initialization and
assembler instantiation. Each test uses its own assembler.
Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Change-Id: I94c996e807b81d8b19361b450be985b12042477c
When using total system memory to test on some motherboard,
the test take very long time, witch looks like system Hang.
Change-Id: Ic31fe60cfb1363fbc8a2d8f7e1cb2bae0e149ea8
Signed-off-by: gaba <gaba@amd.com>
1, Use s_wait_* instead of s_waitcnt
2, Remove a redundant s_waitcnt
Change-Id: Id0f31db0fc520adadd81eb574ad389f63859303a
Signed-off-by: Lang Yu <lang.yu@amd.com>
Add free() for 'all_gpu_id_array' in
hsakmt_fmm_destroy_process_apertures() and
removed it from 'hsakmt_fmm_clear_all_mem()'
Change-Id: I32d2d22e7152f62a3f2e7da4f601f0db7cebd534
Signed-off-by: Apurv Mishra <apurv.mishra@amd.com>
DisableCpQueueByUpdateWithZeroPercentage need to destroy event to avoid
event leak.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Change-Id: I4fb51b670fbff1edcd7fd61517f5c8a6674003c0
The issue arises in the CatchSignal function, which attempts to write to
the standard error stream upon receiving a signal. However, the standard
error stream may already be locked at this point, as the parent process
also attempts to write to the standard error stream after mapping the GPU
memory. This leads to a deadlock, with the program waiting for the
release of the lock on the standard error stream.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Change-Id: Ie69354f4342b96ffe1f2a87f655687da1cbee4b9
there are some timeout issues of evict tests on recent new boards,
it is to solve those issues and optimize evict timeout, as well
as to give user a chance to change timeout in command line.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: I2f40c8ea809c55675b0d0b62296b663481e5fb16
KFDDBGTest and KFDNegative test can eat into memory and event resources
for subsequent test interations if unallocated.
Change-Id: Iea170c20df8d487703441181b6c152b61f02d3db
Queue 2's wave blocked the queue 1's wave save, which will cause unmap
queue preemption fail. Add nop per SQ suggested.
Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Change-Id: Iea7f280e35487059c4499ea999b9e0cdf841d1e1
Current test has 4 processes, each process allocate and access 512
buffers, this requires 2048 waves to access 2048 buffers at same time to
finish the test. For CPX compute partition mode, each compute node has
less waves and cause random test failure. Change test to 2 processes to
use 1024 waves to access 1024 buffers with the increased buffer size.
Add waves_num check to avoid the test failure on new ASICs or simulator,
skip test if the available waves is less than 1024.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Change-Id: I64b5f9172b62cf38f62fbb0b48a801b8a11401c0
This reverts commit 75143555fa.
Reason for revert:
This is currently breaking some tools. Will put it back as soon as tools update their code.
Change-Id: I05c82d443f3a274a618d05e6dc5a87943f5dc7a4
If you build thunk following the instructions in the thunk's README,
there is no /lib folder in the build folder. Adjust the include path,
and clean up the docs to reflect that. The header include is already
defined in the CMake file as ../../include, so we don't use
LIBHSAKMT_PATH for that linking, just the lib location
Change-Id: I73435d59adb9d01f527a28b1935086260e9d3d70
Signed-off-by: Kent Russell <kent.russell@amd.com>
Fixed multiple issues related to memory management, atomicity,
and error handling across various functions: handle null checks,
use-after-free, unchecked returns, and memory leaks.
Change-Id: Ia7c76320cc20e24001052fbba2dd0600bd412140
Currently registering graphics memory without specifying a target
node will return a memory handle that's not a virtual address.
As a result, ROCr is forced to register with a target node for
IPC usage.
Mapping memory without specifying a target node afterwards will
result in mapping to the target node that was imported because the
previous import call flags this node targeting action to future mapping.
For ROCr IPC usage, ROCr wants to map to all GPU nodes if the target node
is not specified.
Allow the caller to register graphics handles that returns a virtual
address without having to specify the target node so that the caller
can make a subsequent map call to all GPUs.
Change-Id: I5a935092b885cc3568e4f3a5dd951c7ec6c84fca
Currently, KFDPerformanceTest.P2PBandWidthTest cannot work if there are
more than 16 KFD nodes in the system. This limit was put in to match the
number of SDMA queues supported on a single node.
This patch updates the test to make it run on systems with more than
16 KFD nodes.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Change-Id: I561d0cdef664cae84fb9c13a801052e2001256e5
In VM with 6vcpu, cpu schedule of
queue_delayed_work(system_freezable_wq) is lower than BM.
HSA_SMI_EVENT_QUEUE_RESTORE event from case HMMProfilingEvent/0 got
delayed execution and caused HMMProfilingEvent/1 fail.
The fix is only listen to HSA_SMI_EVENT_MIGRATE_START event and ignore
all other events.
Change-Id: I534e49b030bd4c534bc7a63eb431f4907659c8cd
Update commentary on HWS scheduler support bit for GPU debugging in
the HSA capabilities node properties field.
Change-Id: I59c519d74a528d5ecf5817ef94e75091314bd844
Fix data race by protecting events_page access with mutex in event create
Fix potential NULL dereference in hsaKmtWaitOnMultipleEvents_Ext
Fix unchecked return value in hsaKmtCreateEvent function
Change-Id: I434bef43666e5205a8b061259569c1d99a952752
We had skipped doing it for PAGE_SIZE, but it should be left as the
regular PAGE_SHIFT name, especially for users who are using different
headers. We want PAGE_SHIFT and PAGE_SIZE to be consistent with one
another, so set them both explicitly to the same value if either
of them is undefined
Change-Id: I121d81c48409dd77351b59a192d824e2419a2410
Signed-off-by: Kent Russell <kent.russell@amd.com>
If BUILD_SHARED_LIBS doesn't get set at all, other projects importing
hsakmt may have an inconsistent state regarding SHARED vs STATIC.
Instead of setting the option and setting it as an option(), just detect
the variable and use it
Change-Id: I9d5a5fc6049ca5351f5e7c63d38ee9bfcb89bdad
Signed-off-by: Kent Russell <kent.russell@amd.com>
The fmm_node_[added|removed] functions were added in the initial FMM
support, but weren't used. Remove them now since no one's referencing
them
Change-Id: I1e46e57294a72012227b38f46c7099de0b9263be
Signed-off-by: Kent Russell <kent.russell@amd.com>
Xiaogang Chen
gaba
Lang Yu
Apurv Mishra
Emily Deng
Eric Huang
Jonathan Kim
Philip Yang
David Yat Sin
Kent Russell
Shweta Khatri
Mukul Joshi
Samuel Zhang