Increasing the timeout will avoid some test failures. This shouldn't
mask any issues as any incomplete shaders should still hang and would
just time out at 180 sec instead of 120 sec.
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: If4e893ab80d9d159bd0b8b112aa7574abc5e4f44
[ROCm/ROCR-Runtime commit: 9168dfe041]
amdgpu_cs_submit can fail intermittently if another process has too much
memory reserved at the time. Allow a small percental of command
submissions to fail to make the test more robust.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: If9f62b2b6f67be71420016d4e38d4dd6b6bca9a5
[ROCm/ROCR-Runtime commit: 8baf02e80b]
Delayed page faults from a terminated process can be attributed to the
next process with the same PASID. Work around that by adding a delay
after the Exception tests to allow the kernel to clean up any fault
storms before the next test.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Id310c13ea9eb92b04d37b95d91a0dd60bd9954e5
[ROCm/ROCR-Runtime commit: bd68646772]
If the signal arrives too late, it interrupts waitpid. Handle this
situation gracefully.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: If4925c352c81ba7fef8a940460b91f5e720b451e
[ROCm/ROCR-Runtime commit: 25288e07dc]
It is gfx90a VF device ID, for virtualization support.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I9e51d6b58c702d185e6758a9c511e9b8bc72c2f5
[ROCm/ROCR-Runtime commit: d8d8e3ddd6]
After unregistered memory is added, now default access attribute
is returned based on xnack configuration.
Change-Id: I8ef44fe1e165ba009622e8112436c1f7a683f6cb
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
[ROCm/ROCR-Runtime commit: 0a2d7d8319]
It is to address gfx90a HW memory model changes.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: Ie5c5c5ee5ddfb75c0b4f625baf59ce37b4cc7c31
[ROCm/ROCR-Runtime commit: a6703395f6]
KFDSVMEvictTest.QueueTest shader asm code need update to support gfx10
and gfx9, skip the test to unblock CI test.
Change-Id: Id2842127cf5fc98a652afa82035a4b3603bf5c33
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 7d53e94750]
Fixes assembler error. The SP3 backend if already set to FamilyId
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: If127a71693b293e2748b06efb668a359b939cd14
[ROCm/ROCR-Runtime commit: e06d549337]
gfx10 GPUs such as gfx1030 need new assembly code to test
the GWS. Removed scalar stores and added proper usage of DLC and
VSCNT waits. Removed gfx9-specific assembler meta-values.
Change-Id: I2bbdb77692ace2dba10997f721ba9decaa9be82a
[ROCm/ROCR-Runtime commit: c1c46d9c97]
KFD changes are ready, all SVM tests should pass now. Skip SVM tests if
the SVM API is not supported.
Change-Id: I5e358565a0458eea45eae0aaf4969ce3a36574a7
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Sierra <Alex.Sierra@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
[ROCm/ROCR-Runtime commit: e8990cf830]
New properties SVMAPISupported added in Thunk spec HSA_CAPABILITY, read
from sysfs from KFD topology.
New local memory property flag CoherentHostAccess added to Thunk
HSA_MEMORYPROPERTY, read from sysfs from KFD topology.
Change-Id: I83933f0e5a61508508168873209dba4af0b77295
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: e8f369b385]
XNACK API for GPUs that support this mode. This API
makes calls to amdgpu driver to configure xnack mode.
It supports set xnack mode and query the current mode used.
Change-Id: If865fd0e3f900f008243dc49504e1a0694e1791a
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
[ROCm/ROCR-Runtime commit: bb441d0bdd]
Add function definitions to support SVM (shared virtual memory)
and xnack set.
Change-Id: Ia97ad9d0c449d8d500d799f702e1a58e87d65a56
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
[ROCm/ROCR-Runtime commit: c44a4be776]
Add svm (shared virtual memory) range and xnack mode
APIs.
Change-Id: Ibd8d7fe566dc200730da0c892caa71aad7589ebd
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
[ROCm/ROCR-Runtime commit: ce26348f3a]
Query the KFD interface version once and store it in a global variable.
This makes it more efficient for KFD APIs to query the API version
later.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I267f3465f754e78fb21a7c42c5877cd68eaa9d05
[ROCm/ROCR-Runtime commit: 43ce63b68b]
This can be overwritten by changing BUILD_SHARED_LIBS=true, but we
default it to static to allow for merging into ROCr
Change-Id: Ic286ef7903a5bc788fe3b84bb13b15bdd3a6f60b
[ROCm/ROCR-Runtime commit: d748d6dce7]
These are failing as well, due to the SP3 shader merge. Blacklist them
as well to avoid more segfaults
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: I07e142a1aad9b2a5304230f333eeaf4392bea4b7
[ROCm/ROCR-Runtime commit: 66da1c9cd2]
The old bit was deprecated, because old buggy user mode depends on it
being always 0. The correct value is now reported in a new bit. New user
mode handles the reported EDC setting correctly, so we can report the
correct value in a new bit.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Ib5d5ed2519810e650458c6b69c97670dab435ddb
[ROCm/ROCR-Runtime commit: d287c60246]
This reverts commit 07b0758bee.
SVM is not ready yet. This was merged by accident.
Change-Id: I8901594a72e785ba5d25a6448718a570e76fe117
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
[ROCm/ROCR-Runtime commit: 41cd7aea2f]
This reverts commit 08e65a397a.
SVM is not ready yet. This was merged by accident.
Change-Id: I1bee102823e7e612be8e8f2e0f50580e8692cc80
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
[ROCm/ROCR-Runtime commit: 5edd00136d]
This reverts commit a247255a6a.
SVM is not ready yet. This was merged by accident.
Change-Id: I372f7d293fd38429ec570bc0e0add7e612871594
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
[ROCm/ROCR-Runtime commit: 4ebda913cd]
This reverts commit 2c1c2cfdf8.
SVM is not ready yet. This was merged by accident.
Change-Id: I7c0d835a0d3a448f2ac1094f818601e5d6363045
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
[ROCm/ROCR-Runtime commit: a8f4c43fef]
BLACKLIST_ALL_ASICS has to the first in the list otherwise "-" negative
flag won't be inserted
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I9ee150d7f793809641b16012929c4e157595d37f
[ROCm/ROCR-Runtime commit: 0f16e6f35f]
Possibly because of moving to gart table for vram access from Kernel.
This test failure shouldn't be a blocker. Temporarily blacklist till a
solution is found.
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I99725f368aced863188e30f619288ad4d033b9a6
[ROCm/ROCR-Runtime commit: e35778ed4d]
Set the KFD_IOC_ALLOC_MEM_FLAGS_COHERENT flag and
KFD_IOC_ALLOC_MEM_FLAGS_UNCACHED flag to allocate
uncached coherent memory when HSA_DISABLE_CACHE
environment variable is set. At KFD driver,
Single KFD_IOC_ALLOC_MEM_FLAGS_UNCACHED flag is
not sufficient to allocate uncached memory. We
have to use both two flags to allocate uncached
memory.
Change-Id: Ie490f37b2e696314e60048f5b1b57442431696e9
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
[ROCm/ROCR-Runtime commit: ae0e74095e]
Three kfd subtests are added to verify new XGMI connection with
cache coherence HW link on A+A.
Signed-off-by: Eric Huang <JinhuiEric.Huang@amd.com>
Change-Id: I6960ec91cbfb696c4e6acb3b79fd83107003acdd
[ROCm/ROCR-Runtime commit: 9aa521d1ff]
In A+A all system memory is mapped as NC. So add a new function
gfx9_PollNCMemory which will support NC memory.
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I097b95fb156f73d6f480cd4fd262cc6fa5933f69
[ROCm/ROCR-Runtime commit: 085005f07b]
destBuf is mapped as cached, the intruction flat_atomic_add
operates on cache that cause test failed. Adding scc modifier
in the instruction will fix the issue.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: I8e138f93ae4f5e23020e3ac1549ef924968a74c5
[ROCm/ROCR-Runtime commit: f7759df6e0]
The three kfdtests have been fixed, so remove them from
filter list.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: I101a72476970a9d105e8c0b5c022847757fdd316
[ROCm/ROCR-Runtime commit: 3a378fcf0b]
Before gfx90a, coherent memory is uncached. So it was reasonable
when environment variable HSA_DISABLE_CACHE is set, memory is mapped
as coherent. On gfx90a, coherent memory can be cached, so mapping
memory as coherent can't guarantee memory is uncached. When
HSA_DISABLE_CACHE is set, we have to map memory as uncached.
Change-Id: Ia5ed4cf0ad6aef5644dc8c9e6632b52d606f06f4
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
[ROCm/ROCR-Runtime commit: f132fb2cd0]
Refer to commit "Mark buffers accessed by CP as UC"
A+A buffers are mapped as NC. CP (PM4Writes) need ReleaseMem function to
ensure the write go through to the memory
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I4ee55a6e40fba078f5950d95c8fee7ee076260bf
[ROCm/ROCR-Runtime commit: 57f46b53ec]
Refer to commit: " Mark buffers accessed by CP as UC"
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I1816e035dbb3178f28f5e34b050c20ecca282060
[ROCm/ROCR-Runtime commit: 0e8500b886]
This change might be redundant if ROCr takes care of it
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I7b67143a8ad21baa61b7eda7b8e5fe0ac1e33830
[ROCm/ROCR-Runtime commit: 10674916e4]
This change is for the A+A bring-up branch as it needs to made more
generic to handle all ASICs.
For A+A all the system buffers are mapped as NC (non coherent) unless
explicitly marked as UC (uncached). The coherency is then expected to be
handled by shader by explicitly using acquire/release instructions.
However, CP doesn't have same feature. The buffers used by CP thus have
to UC. For now queue buffer and Signal handler memory is marked as UC.
This change shouldn't affect other ASICs since Uncached flag is not used
in those. However, this change still need to be made more generic.
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: I56c37a809913f7f08c94d01b0572d0f4864939aa
[ROCm/ROCR-Runtime commit: 7c05c5240f]
On gfx9, the maximum number of wavefronts per queue is the minimum of
40 waves per compute units, or 512 waves per shader engine. On gfx10,
there can only be 32 waves per compute units.
Signed-off-by: Laurent Morichetti <laurent.morichetti@amd.com>
Change-Id: I148d1a4fe6c07cdbfaa1f77939eb29311c81c008
[ROCm/ROCR-Runtime commit: 4cf11c3a7e]
Reserve some space in the context save area for the debugger's
use. There should be 32 bytes per wave for a given queue.
Change-Id: I65ddb6123d0f6afd3149844617ad19023009101d
[ROCm/ROCR-Runtime commit: a83f9b67ce]
Temporarily blacklist some tests on gfx90a until they are solved.
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Change-Id: I87cc3a996ea7d55ed8f20f5b4eecfd8bb691effd
[ROCm/ROCR-Runtime commit: e342c9c890]
On every new asic with new stepping, we need to manually relax this
checking. This check is not very helpful. Delete it.
Change-Id: I11f813023ca2566d82f6d11121d4be38c296674b
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
[ROCm/ROCR-Runtime commit: 1f05b54dc9]
Due to cache coherence change, the remote vram mapping is changed
to cached, the written value by remote shader will not be read by
local shader. So the test will fail.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: I2b64e8a30bed0066e159bad9bb7febae5ebe84aa
[ROCm/ROCR-Runtime commit: ec7ba38b23]
Kent Russell
Mike Li
Felix Kuehling
Alex Sierra
Harish Kasiviswanathan
Eric Huang
Philip Yang
Joseph Greathouse
Roberto Di Remigio
Cole Nelson
Oak Zeng
Chengming Gui
Laurent Morichetti
Amber Lin