Detect under-reporting of available memory by initially attempting to
allocate substantially more than reported available memory, and ensure
that the allocation fails. Continue shrinking the attempted allocation
until it succeeds, then fail the test if the successful allocation is
either too much more than or too much less than reported available.
Signed-off-by: Daniel Phillips <daniel.phillips@amd.com>
Change-Id: Ib418f0aa26e8db80590a6c5f2578da56a4b60f2b
[ROCm/ROCR-Runtime commit: e71eb13784]
Modifier scc is disabled from gfx90a's asm, so remove the
shader for gfx90a A+A and keep it for newer asics with scc
support.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: Iec3c7ccd5156a855adb2b02feb3db0761876aa2f
[ROCm/ROCR-Runtime commit: 8e8aa024fd]
To avoid confusion since this shader has changed to be persistent
(original IterateIsa may be re-used for debugger tests).
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I4643692765fc7665933257e89d5b922e779ad2e5
[ROCm/ROCR-Runtime commit: 6467664ec7]
Get more debug information about user pointers that were registered
through SVM API, and triggered by memory exception events.
A new kfdtest with this use case was also included inside
KFDExceptionTest.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: I8e9df3c1c6c3f42d7b9235d12406d80d31746443
[ROCm/ROCR-Runtime commit: 45fad29752]
Register and map userptrs through Shared Virtual Memory(SVM) API at
the Kernel level when available. Using this approach, performance
will be improve as register/unregister memory will not trigger any
system call to KFD driver.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: I20723cbeb340bf48b95e1115f0102c031397bc14
[ROCm/ROCR-Runtime commit: 178a619b80]
KFDCWSRTest.BasicTest is parameterized to allow an easy method of
tweaking the number of work-items (and save/restores). The input/output
buffers were previously hardcoded to a single page, which would cause a
segmentation fault if the number of work-items specified is greater than
1024 for wave32.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Ieefc819a5d81c77cee88081a287fd383e6378e74
[ROCm/ROCR-Runtime commit: 73adbdee2c]
For software trap in GFX11, COMPUTE_PGM_RSRC1 must have PRIV = 1.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Id504889c3ca2588b6c8cefdebaec00dcfc217995
[ROCm/ROCR-Runtime commit: 6294ef564b]
On error mmap returns value MAP_FAILED, which is (void *)-1, not NULL
pointer.
Change-Id: I81b187266c943fa0aa4fab21b529d4c2989b12ad
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 590fd531c0]
IterateIsa had some leftover instructions from when the shader was
getting updated for KFDCWSRTest.BasicTest.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I41ae7b7948cbe2aff8bf61b170b9a7d498b836a3
[ROCm/ROCR-Runtime commit: 82a41c7e4d]
When running kfdtest test case, because the filter node of the new chip is
missing in libhsakmt, the test case is not supported, so a new test node
is added in order to spporting kfdtest case.
Signed-off-by: shikaguo <shikai.guo@amd.com>
Change-Id: I0cd9ffd7d4387129cfb0f8de6b669f431949ab49
[ROCm/ROCR-Runtime commit: 4951495fca]
Required due to LLVM retirement of llvm::apply_tuple, instead using
std::apply which was introduced in C++17.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I6646ebcca7d71d3e1bcf340ccfa3db2c15a3110a
[ROCm/ROCR-Runtime commit: 4267c4b524]
Failure with new CWSR tests reported for GFX10, for now add to blacklist.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I5b2bd9ec61c64ad66e1c34ba2c192bece808f56f
[ROCm/ROCR-Runtime commit: 0055ef46c4]
This patch restructures the CWSR basic test and allows for
creating parameterized CWSR tests. This patch introduces four
parameterizations. These tests behave as follows:
This test dispatches the IterateIsa shader, which continuously
increments a vgpr for (num_witems / WAVE_SIZE) waves. While this shader
is running, dequeue/requeue requests are sent in a loop to trigger
CWSRs.
This test defines a CWSR threshold. Once the number of CWSRs triggered
reaches the threshold, a known-value is filled into the inputBuf to
signal the shader to exit.
4 parameterized tests are defined:
KFDCWSRTest.BasicTest/0
KFDCWSRTest.BasicTest/1
KFDCWSRTest.BasicTest/2
KFDCWSRTest.BasicTest/3
0: 1 work-item, CWSR threshold of 10
1: 256 work-items, CWSR threshold of 50
2: 512 work-items, CWSR threshold of 100
3: 1024 work-items, CWSR threshold of 1000
Tuple Format: (num_witems, cwsr_thresh)
num_witems: Defines the number of work-items.
cwsr_thresh: Defines the number of CWSRs to trigger.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I639eb7bd75b14ee70e190b4bd19dcf34096fc7bf
[ROCm/ROCR-Runtime commit: 0dbac97b75]
This didn't return anything, so add a "return 0" at the end, since the
function expects to return an int value
Change-Id: I17c398e431b2ce4571e6ca4abe6d567f110ea2a7
[ROCm/ROCR-Runtime commit: 90ada94141]
: The kernel driver will do align VRAM allocations to 2MB, instead of 4KB.
Change-Id: Iea9d8c0f02999b9ea5fd931da82240a33f7bcc69
[ROCm/ROCR-Runtime commit: 17fb40f1f6]
KFDExceptionTest.SdmaQueueException allocates VRAM with host access. This
fails on small-BAR GPUs. This error was incorrectly ignored before
7ccda4ba26 ("kfdtest: Full TearDown and SetUp in child process").
The test doesn't really need host access to the memory. Therefore the fix
is to disable the HostAccess flag.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Ifec279eeb6c1ecb1160db9b692e6dc8816d761a3
[ROCm/ROCR-Runtime commit: 9d33827a84]
The CMA feature is deprecated and about to be removed from the DKMS
branch. It was never supported upstream.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I64b6213eb3adbdc550542e51181cd8ba6ca4cb45
[ROCm/ROCR-Runtime commit: cdaaf8236a]
With the next patch, child processes need to fully reinitialize the
topology in order to recreate the process apertures. Just calling
hsaKmtOpenKFD is no longer sufficient. Tests based on
KFDMultiProcessTest already did this correctly (KFDHWSTest, eviction
tests). This patch fixes KFDExceptionTest and KFDIPCTest.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Iaad24e88ddd29c1105bf791a77891cc55a6072ff
[ROCm/ROCR-Runtime commit: 412b24137e]
LoopIsa is a shader that performs a variety of intensive
calculations in a loop. It is used by tests such as
KFDQMTest.QueuePriorityOn*
It contained a scalar load, despite not having any buffer to
read from. This load causes page faults on GFX11. It is
unclear why it did not cause page faults on earlier ASICs.
Remove the load.
Signed-off-by: David Francis <David.Francis@amd.com>
Change-Id: I7426d0db48e933f3bb870467ea88476f7a283040
[ROCm/ROCR-Runtime commit: 39e8a85aac]
When the shaders were moved to ShaderStore,
KFDQMTest.EmptyDispatch was erroneously
changed to use LoopIsa instead of NoopIsa.
Change it back.
Signed-off-by: David Francis <David.Francis@amd.com>
Change-Id: Iaf7d0d107e3bf3bd8b7d616b137a1740e309cf91
[ROCm/ROCR-Runtime commit: 4b8c74bf04]
Previously we omitted the version and arch in the filenames. By adding this,
as well as the ROCM build variable, this will allow for easy version
version detection on systems. Instead of kfdtest = v1.0.0, now it will
feature the build number, allowing for easier identification as to which
version is installed.
Change-Id: I311ed7010486e7c70af669d282910fe29ee8db45
[ROCm/ROCR-Runtime commit: 9745db3053]
Open SMI event file handle, prefetch to migrate svm range to GPU, read
HMM profiling events, then check event_id, address, size, pid, event
triggers are the expected value.
Start separate thread to read SMI event, the same way applications use.
Use thread barrier to ensure no event is dropped.
Change-Id: I0683969d18d1579847e125d86aa4257602adb13f
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 7799611c01]
Kernel amd-staging-drm-next branch changes GFX11 fish_colour sysfs
naming to "ip discovery". Update run_kfdtest.sh to use sysfs
gfx_target_version for ASICs that have transitioned to IP discovery
topology.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: If202a0ceeed7324364539a33661f0abcf0973f07
[ROCm/ROCR-Runtime commit: 350eba3a07]
Currently, using --gtest_filter will override any exclusions from
kfdtest.exclude. To add an additional test (or set of tests) to the
exclusion list dynamically, the --exclude (-e) flag will allow the user
to pass in a string of GTest-style exclusions to add to the list
generated by kfdtest.exclude
e.g. run_kfdtest.sh -e "*KFDLocalMemoryTest.*:KFDEventTest.*"
will use kfdtest.exclude, but will also exclude all LocalMemory and
Event tests
Change-Id: Ic23ec271ba2cd2240d2e98558c0117ff2a064ed2
[ROCm/ROCR-Runtime commit: f7978b1ff6]
Add blacklist for tests on GFX11 that are under debug/not functional.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I32f375f29036a2fc1a54f2792b31ebc45f4d668b
[ROCm/ROCR-Runtime commit: 852d29bca2]
GFX11 set to FAMILY_NV.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I545241c39bbe739e39a8943b242b9fc49a65a7e1
[ROCm/ROCR-Runtime commit: c198b91c94]
With this patch gfx11 is supported in amd-staging.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I4bbfc87345a486dd8f2e0091ea7b82c255a8ad15
[ROCm/ROCR-Runtime commit: cee64d4825]
Old CMake message was misleading, update to reflect what it's actually
doing.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I8fd18a9095ea350cc10494b4382706027b447ba3
[ROCm/ROCR-Runtime commit: d5f074902c]
KFDQMTest.BasicCuMaskingEven was previously in TEMP_GFX10_BLACKLIST and
was later added to BLACKLIST_GFX10_NV2X. shows this test
also has problems on NV1X, so move to BLACKLIST_GFX10 to blacklist for
all GFX10 until resolved.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I75426780a52172ed0f2468f0297241902da5002a
[ROCm/ROCR-Runtime commit: 30bba89830]
Add BLACKLIST_GFX10 and BLACKLIST_GFX10_NV2X to reduce redundant
individual blacklists. Remove KFDDBGTest from kfdtest.exclude and
CMakeLists.txt as it is deprecated.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I5ffebb012ff8ef4bba5aadfeb3c7478fe43c65f1
[ROCm/ROCR-Runtime commit: 0955f350e3]
DeviceHdpFlush was part of the prior TEMP_GFX10_BLACKLIST and is
soft-hanging after being re-added to the test list. Temporarily move
back to gfx10 blacklists until root-caused ().
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Id37b7105566e8243465e4b21ebc2e2e8c4c79923
[ROCm/ROCR-Runtime commit: fe2098e011]
In DeviceHdpFlush, isaBuffer was accidentally used instead of isaBuffer0
during LLVM re-work--revert.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I98d2322e772f821a39505bb336ceb4e6cd8722ef
[ROCm/ROCR-Runtime commit: ffa6c95858]
Otherwise the test failed because KFD returns failure to map or prefetch
the SVM range to those GPUs which don't support SVM.
Change-Id: Ideaec2a686bfabb2fd6ae41b1a690da144121a89
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: c05c66e756]
In the LLVM rework, a line was accidentally changed from
isaBuffer1 to isaBuffer, causing VramCacheCoherenceWithRemoteGPU
to fail. Change it back.
Signed-off-by: David Francis <David.Francis@amd.com>
Change-Id: Ie1f3465b5c46556f18682d1b3d1f086bb790c648
[ROCm/ROCR-Runtime commit: 4b041a8ad9]
[WHY]
These tests force HW exceptions in the GPU driver. Some of these
exceptions might print page fault error messages at kernel level.
These are expected errors due to the nature of the tests, but still
might cause confusion to users.
[HOW]
Add log message to warn the user about these kernel error messages.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: I8eef87b83939e37230da0c374c2f77d2d484baa9
[ROCm/ROCR-Runtime commit: b9e8bc1f52]
Includes a simple AssembleShader test which loops through all shaders
for all supported targets, dispatching a RunAssemble call for each
shader.
Also adds extra safety on a couple shaders that only work on
gfx9/gfx90a.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I3ca1c92136f3871eb62fcb9645694f22287aaeec
[ROCm/ROCR-Runtime commit: 7eeba830f8]
With LLVM-based assembly these shaders are now valid for GFX10, with the
exception of KFDSVMEvictTest.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Idc872139176bbc1cc8d7ae61a8e4572360ecb5d5
[ROCm/ROCR-Runtime commit: 025c6146d9]
Daniel Phillips
Eric Huang
Nirmal Unnikrishnan
Graham Sider
Ramesh Errabolu
Alex Sierra
Yifan Zhang
Philip Yang
Daniel Phillips
Shikai Guo
Kent Russell
jie1zhan
Felix Kuehling
David Francis