Add HsaMemFlags Contiguous bit for hsaKmtAllocMemory to allocate
contiguous VRAM, to support RDMA device with limited scatter-gather
ability.
Check KFD ioctl minor version >= 17.
Change-Id: I0db00dad125b2b7be523f343082641f59b850423
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 97497c7efc]
HsaMemFlag new flags added and the number of the reserved bits is
reduced, and generate value overflow compilanation error.
The reserved bits is not used, remove the init.
Change-Id: I603596977dfd558ce31ead03711d7c5ce5ee5b71
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: e2d742ac6f]
The build tree was missing a level of nesting, causing diversions based
on in-tree/out-of-tree use.
KR: Also fixed kfdtest paths
Change-Id: I8638b6d6227daabddd8eaa2aa387ba578b8dfab8
Signed-off-by: Stella Laurenzo <stellaraccident@gmail.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
[ROCm/ROCR-Runtime commit: a180fea5ad]
When allocating memory for MES AQL queue structure, the PreferredNode
is set to the device index of GPU to hint the location where the BO
needs to be created. But we need to ignore the device index when calling
bind_mem_to_numa.
Change-Id: Iae69fe02bfd48c5a3bd495319f6f2706d6e8aea2
[ROCm/ROCR-Runtime commit: 541d0dbbae]
Add pc sampling support.
Change-Id: I08199024ba5a8eb2845c048d499fc8fcd260d2e8
Signed-off-by: David Yat Sin <David.YatSin@amd.com>
Signed-off-by: James Zhu <James.Zhu@amd.com>
[ROCm/ROCR-Runtime commit: f94e2530fb]
To allocate GTT memory for MES AQL queue structure, KFD will create GART
mapping for the memory to be accessed by MES.
Change-Id: Iae7b33d1e70861109f1551d3a71dc60dfde9de61
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 9fbe853fea]
This allows to build ROCT-Thunk-Interface for Alpine Linux, Gentoo with musl profile and so on.
List of changes:
* Fix redefinition of PAGE_SIZE from limits.h
* Use NAME_MAX from limits.h
Closes#65
Change-Id: Ibdb0ef5668a07b7b403fcc4a44cd2658e00a584a
Signed-off-by: Sv. Lockal <lockalsash@gmail.com>
Signed-off-by: Kent Russell <kent.russell@amd.com>
[ROCm/ROCR-Runtime commit: 9a89997b5f]
Pass the correct gpu_id to KFD for system memory that is allocated for
the queue and eop buffer
Change-Id: I43bb6333560a7d9d38293c191303161ab1443b5d
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
[ROCm/ROCR-Runtime commit: 341ecaf1d9]
For GTT memory allocation if GPU is provided honour it.
Change-Id: Iea9a26bc44cd3daa2337845f53dc430787b0643b
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
[ROCm/ROCR-Runtime commit: 858dfd364f]
When HSA_USE_SVM is 0, thunk uses non-SVM path, but upper layers still
use SVM path. That is not as expected.
Suggested-by: Lang Yu <Lang.Yu@amd.com>
Change-Id: I1ae0b4faa2f8af5ec69a81cfeb7661bd47d739d4
[ROCm/ROCR-Runtime commit: 0accd17b6e]
If using hsakmt as a shared library
Change-Id: I66a1849a46bd7009813d49824d0d059e8a511038
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
[ROCm/ROCR-Runtime commit: 42581d4172]
For "Intel Meteor lake Mobile", the cache info is not in sysfs,
That means /sys/devices/system/node/node%d/%s/cache is not exist,
but system working fine.
Change-Id: Ie7c04426791a84c2288ff21df093226828a5f629
Signed-off-by: Gang Ba <Gang.Ba@amd.com>
[ROCm/ROCR-Runtime commit: 4bf73f521b]
Add new structures for HW Exception events and copy data from KFD to
expose to upper layers.
Change-Id: Icd5eb98997c47620e3b86277ab6d3abb7ed7d56f
[ROCm/ROCR-Runtime commit: 01ff2f7934]
For large memory allocations (>2MB) the thunk should use the
MADV_HUGEPAGE flag for madvise call to optimize allocation performance
on certain operating systems that rely on madvise hint when Traspatent
Huge Pages is not set to always.
Suggested-by: Joseph Greathouse <joseph.greathouse@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Change-Id: Ic0c753f89a177b0f715942d6e2a7108b08a85f20
[ROCm/ROCR-Runtime commit: 5047eb161f]
Set CWSR svm range granularity to 0xff, then KFD will migrate the entire
CWSR range from VRAM back to system memory when recovering the CPU page
fault if rocgdb access CWSR area, this avoid the partial CWSR range
migration and stall CWSR GPU mapping issue.
This is a temporary workaround, it should be reverted once the KFD is
fixed.
Change-Id: I80a7248244574edba25b13858b7ebcf1c77b8930
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 85a47fa66b]
This share resource is for IOMMUv2 which is removed from
AMDGPU/KFD.
Change-Id: Ia6e9311f1adc56fac2c9e8fa05b24c5ec8c272a5
Signed-off-by: James Zhu <James.Zhu@amd.com>
[ROCm/ROCR-Runtime commit: d195deeec4]
IOMMUv2 is removed from AMDGPU/KFD.
Change-Id: I9fcf20ae9288cb40bb4b696284fc70534fb6484b
Signed-off-by: James Zhu <James.Zhu@amd.com>
[ROCm/ROCR-Runtime commit: 277d5e27ff]
IOMMUv2 is removed from AMDGPU/KFD.
Change-Id: Ib87f501c07d9de90e6b83b98f98daacd5913e98a
Signed-off-by: James Zhu <James.Zhu@amd.com>
[ROCm/ROCR-Runtime commit: 274b5b51ca]
Add support for new flag for memory allocation that will provide
system-scope coherent atomics
Change-Id: I426d66223e8d2b570f69b4c0e61145ce9b2290d2
[ROCm/ROCR-Runtime commit: 8e06dce573]
Add base debug operations to suspend and resume queues.
Routine will return the number of queues successfully
suspended or resumed.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Change-Id: I8f18317f70464b04231c5cf822e11d545ebfa02a
[ROCm/ROCR-Runtime commit: 5a675921ea]
Add base debug class and attach/detach operations.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Change-Id: I60f3c166646f05838fec208ac2f59bba998c63f8
[ROCm/ROCR-Runtime commit: dd56b38c2f]
Even if the version of libdrm older and does not support the
amdgpu_device_get_fd function, the device_handle stored in
amdgpu_handle[] is still valid and can be returned via
hsaKmtGetAMDGPUDeviceHandle.
Change-Id: I024a3e82e6cfebac5577aefe359b067746c4023e
[ROCm/ROCR-Runtime commit: 66b66e42cd]
Functions to API added to extract the following information from KFD
Runtime information, device info and queues snapshot.
Signed-off-by: Alex Sierra <Alex.Sierra@amd.com>
Change-Id: If995ecc54497ab61189bb0f209c64af0bbb0f56f
[ROCm/ROCR-Runtime commit: 0cbf26c148]
Queries for runtime capabilities after its being enabled
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: I098c0e9862c0c1d5e304b111cdc281c0ccd09691
[ROCm/ROCR-Runtime commit: 5e0a32d7b3]
Keeping last signaled event age to avoid race conditions
for HSA_EVENTTYPE_SIGNAL when event age init value is non-zero.
Change-Id: Ifb9a11a6868e5762a9f92f579e45a0a2c8fa1017
Signed-off-by: James Zhu <James.Zhu@amd.com>
[ROCm/ROCR-Runtime commit: a0cbf90b90]
If we end up in the first if clause, aperture_base is not set, unlike
the other 2 clauses. Initialize it to NULL at declaration time, and only
change its value in the final else clause, where we set it to
aperture->base
Change-Id: I2bf44dc93cae8a03e66f41cedd85d57be2115bba
Signed-off-by: Kent Russell <kent.russell@amd.com>
[ROCm/ROCR-Runtime commit: 718d95de77]
Allow hsaKmtRegisterGraphicsHandleToNodes parameters NodeArray be null
and NumberOfNodes be zero at same time. It is the case we want the imported
buffer not be registered by kfd. Set gpu_id_array = NULL explicitly to avoid
free uninitialized gpuid array.
Report: Yat Sin, David<David.YatSin@amd.com>
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I3babc1160c9573e38dd11d81965c8de2b70cae2e
[ROCm/ROCR-Runtime commit: f6183f937e]
Have hsaKmtMapMemoryToGPU return same value as fmm_map_to_gpu to keep consistency.
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: Ifabb72301e1d5a6c1310973bb1321714e12a1fa6
[ROCm/ROCR-Runtime commit: 7e4e57ae5f]
Query render node fds that libdrm uses for current process and
use them at Thunk if available.
v2: avoid naming conflict with amdgpu_device_get_fd from amdgpu.h
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: Id7288c03730f4a4c9c3644e37ca4725fec71a471
[ROCm/ROCR-Runtime commit: ac1db60fc2]
When gpu map info is not provided import DMABuf without VA assigned.
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I996ab4eb46977af5064126529c28a8bf20a67292
[ROCm/ROCR-Runtime commit: 989c6c617c]
This new manageable_aperture_t is used for VRAM allocation-only and
VA allocation-only.
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I3866ef9d35386d6aef7b6934ac8d4a89ef843b50
[ROCm/ROCR-Runtime commit: 0a2989083b]
This reverts commit 89ce41694f.
Current amdgpu exposes one render node for one gpu node/partition,
revert to previous way to open render node at Thunk.
Signed-off-by: Xiaogang Chen<Xiaogang.Chen@amd.com>
Change-Id: I436be74f8e872a7ab5c4a1420b4ea884f5a00e57
[ROCm/ROCR-Runtime commit: cc4fb2d1a9]
Add support functions to remap the first page of device memory (GPU/GTT)
to share host ASAN logic.
Signed-off-by: David Yat Sin <David.YatSin@amd.com>
Change-Id: I4c27d5417ba80a172dccb0a079a597c5dc1c8f85
[ROCm/ROCR-Runtime commit: 1e6d728730]
Previous code might fail to get the correct ln node. And trigger extra
walk through of the tree. Fix it.
While walking through the tree, better to search from right to left as
the node->start likely close to *address*.
Change-Id: If86ddf73e59a1eb88225d1ea90797818e8165488
Signed-off-by: xinhui pan <xinhui.pan@amd.com>
[ROCm/ROCR-Runtime commit: 77761836ae]
We should check compute core instead of cpu core,
in order to exclude the case of APU.
Signed-off-by: Jesse zhang <jesse.zhang@amd.com>
Change-Id: I2ec2a6807f51f49f80e0e500f5d9af81c2efae37
[ROCm/ROCR-Runtime commit: 4d54d6e706]
GC 9.4.3 to set gfx target version to 9.4.x dependent on revision and
capabilities. Due to this, where applicable, mask off the gfx target
stepping version and only check major/minor version (9.4). There are no
collisions due to this change since GC 9.4.3 is the only ASIC that uses
gfx target version 9.4.x.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I72803e594c421f054d18ccfa7e92c507128fa5be
[ROCm/ROCR-Runtime commit: 831d1ad352]
Use q->total_mem_alloc_size for munmap in SVM codepath of free_queue.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: I2fecaa1ddb337b1fe71f9cbba45a0c9467eff0c0
[ROCm/ROCR-Runtime commit: ae659e5427]
Currently, on queue destroy, context save restore memory is freed
only for a single XCC. Instead, we need to free the entire context
save restore memory, which was allocated for all XCCs.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Change-Id: I51ebb12fa8d5ebed41979d68e74f7c5392dca062
[ROCm/ROCR-Runtime commit: a713fb766e]
Do not allocate the EOP buffer when not required.
Signed-off-by: David Belanger <david.belanger@amd.com>
Change-Id: I1664a3f0a882219a72278174006cdb8d46fd4f5e
[ROCm/ROCR-Runtime commit: 252a2cf959]
Philip Yang
Stella Laurenzo
David Yat Sin
Mihai Preda
Xinmudotmoe
Sv. Lockal
Harish Kasiviswanathan
Jeremy Newton
gaba
Rajneesh Bhardwaj
Torsten Keßler
James Zhu
Jonathan Kim
Alex Sierra
Kent Russell
Xiaogang Chen
Bing Ma
xinhui pan
Jesse Zhang
Graham Sider
Mukul Joshi
David Belanger