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]
In multi-partition modes, e.g. CPX, we want to create new file
descriptor despite using the same render node. Update
open_drm_render_device to use a gpu_id to fd map partitioned by render
node. Different gpu_id's requesting the same render node will be added
to that render node's map list for fetching its fd. Different gpu_id's
requesting different render nodes as well as the same gpu_id's
requesting the same render node will behave as they did previously.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Change-Id: Ie153d42355d4d75b1c6ba6ff40fac3295bc87009
[ROCm/ROCR-Runtime commit: fd48f14ceb]
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: I3726b4b5e1c6a52a83786fbe0af6322eb29ae7c9
[ROCm/ROCR-Runtime commit: 63c8cf115a]
Implement hsaKmtExportDMABufHandle, which can be used for a new
upstreamable RDMA solution. It exports a DMABuf handle for an arbitrary
virtual address along with the offset of the address within the
allocation. It also checks that the size of the intended export does
not exceed the allocation.
This uses the new AMDKFD_IOC_EXPORT_DMABUF, which requires KFD ioctl
API version 1.12.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Ie5fdb1f73ab3c7fa36c315ce326b1fb89eacc8b6
[ROCm/ROCR-Runtime commit: 332f59eb2a]
This reverts commit 1bb6d872ac.
It causes a regression in pytorch benchmark.
Change-Id: I96173dbd061cf38d6f451c02cb181ae51b7f625e
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
[ROCm/ROCR-Runtime commit: 505287412f]
This reverts commit ea19fbb646.
There are some openMP issues that were introduced after SVM userptr
feature was added.
Signed-off-by: Alex Sierra <Alex.Sierra@amd.com>
Change-Id: I7ef87c5232a3bcbe594c743fa4b4958601845ba5
[ROCm/ROCR-Runtime commit: f2bda56d04]
This reverts commit 6789a0f3bd.
There are some openMP issues that were introduced after SVM userptr
feature was added.
Signed-off-by: Alex Sierra <Alex.Sierra@amd.com>
Change-Id: Ib01046571d2c84fa0fd228ecba0dee0eae3f994d
[ROCm/ROCR-Runtime commit: 21e95a4f2a]
This is handled by __fmm_release calling aperture_release_area.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Ib8ed300e1734f03aeb9dfc8074897ece310b8af9
[ROCm/ROCR-Runtime commit: 7787a039bd]
Use a common helper for CPU mappings to reduce duplicate code.
Consistently use MAP_SHARED for all render_fd mappings.
Remove double-mapping for AQL queue buffers on the CPU. This workaround
is only needed on the GPU.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Iff86c8cc9f1e5c982614b3f11129bc2cf8cbba02
[ROCm/ROCR-Runtime commit: 73b0fb3d7c]
The NULL pointer check was the only way for that function to fail. And it
was done after the pointer was accessed. Simplify this by just returning
the result as a return value instead of using a pointer as output
parameter. This way the function can never fail and the caller doesn't
need to do any error handling.
Declare the function in libhsakmt.h instead of duplicating the
declaration in fmm.c.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I91b90d66166fd3b5cdc47c73a9bbc369c45b51fe
[ROCm/ROCR-Runtime commit: 2d53430ce3]
Setting this variable to '0' will force to disable memory
registration/allocation through SVM API mechanism.
Not setting this or setting to '1', SVM API will be used only if all
GPUs support it.
Signed-off-by: Alex Sierra <Alex.Sierra@amd.com>
Change-Id: Icdf7656de09aa9988b567ec6c024953398e9bb48
[ROCm/ROCR-Runtime commit: 8a746bdaed]
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]
fork process copy-on-write MMU nitifier on CWSR range will evict user
queues, and then update GPU mapping and resume queues, use MADV_DONTFORK
to avoid COW MMU notifier callback on CWSR SVM range.
Use mmap to alloc SVM range for CWSR because posix_memalign don't alloc
new range in child process, this fails to register svm range as range is
invalid address in forked child process.
Change-Id: Ibaea56a691dd6f577ed2e1f2d43f4a3500b8316f
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 093cf898fb]
mmap alloc larger address range with align padding page plus guard
pages, then unmap the padding and guard pages at beginning and end
of the range, return aligned address range.
Change-Id: Iaf3c711a079c744289efbafee9b5e63aaf724765
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: b7710a1dda]
Avoiding segfault when an invalid SharedMemoryHandle is passed in
when calling fmm_register_shared_memory.
Change-Id: I0e0bbed01487fc10afcbb170eb9330e70b209d14
Signed-off-by: David Yat Sin <David.YatSin@amd.com>
[ROCm/ROCR-Runtime commit: 1c385fb257]
hsaKmtMapMemoryToGPU should not try to map VRAM on peer GPUs that don't
have an IO-Link to the memory. The new P2P mapping code in KFD will
fail otherwise.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I6d59b55651b98756865a0f69eafef3e386372cf3
[ROCm/ROCR-Runtime commit: 9ac2c75171]
This allows init_process_apertures to use the whole consistent topoology
instead of taking its own partial snapshot.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: Ia13e7aa7fcd090ea8d6cacd4babb29a27c20207f
[ROCm/ROCR-Runtime commit: 87aca673e8]
Add KFDGpuID to HsaNodeProperties to return gpu_id to upper layer,
gpu_id is hash ID generated by KFD to distinguish GPUs on the system.
ROCr and ROCProfiler will use gpu_id to analyze SMI event message.
Change-Id: I6eabe6849230e04120674f5bc55e6ea254a532d6
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 88778e13dc]
This env variable sets the max VA alignment order size as
"PAGE_SIZE * 2^alignment order" during mapping. By default the order
size is set to 9(2MB).
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: I01ae4e0963f4d21c7c367464e60f865bc58d7fac
[ROCm/ROCR-Runtime commit: 0e908f05bb]
Tested on Talos II with Vega 64
POWER systems allocate NUMA nodes on multiples of 8 to allow CPU
onlining / offlining
Set the correct NUMA mask bits when requesting node-bound memory
allocations
This is a cleanup/squash/rebase of:
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/pull/47
Change-Id: Id4af6dff7e66e9d464d6b17a1e99087eb3ac8e51
Signed-off-by: Jeremy Newton <Jeremy.Newton@amd.com>
[ROCm/ROCR-Runtime commit: 5fd3c868b2]
Mapped memory areas become invalid after fork, and the child process is
required to remap the memory areas after a fork. So we mark these device
memory mappings with MADV_DONTFORK so that they are removed from the
child process after fork.
This was causing some issues when doing CRIU checkpoint/restore because
CRIU and amdgpu_plugin were not able to handle these mappings.
Change-Id: I50eb334aecea6dab7522d94da0273adcf4fb1ce0
Signed-off-by: David Yat Sin <david.yatsin@amd.com>
[ROCm/ROCR-Runtime commit: 4986f4a5c2]
For userptr, after taking aperture lock, decrease registration_count and
ensure object registration_count equal to 0 to release KFD and thunk
object.
Move decrementing of registration count from fmm_deregister_memory into
__fmm_release to avoid a race condition when dropping the aperture lock
in fmm_deregister_memory.
Change-Id: I5381fa6b8a77a1516af2554e5174e91969c338c4
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: f82927ad65]
After calling ioctl to create userptr obj, take aperture lock, check if
there is same userptr obj created after finding object, to catch the
race that multiple threads register same userptr to multiple GPUs.
If same userptr obj exist, then increase userptr registeration_count,
and free the newly create obj.
Change-Id: I63ae3a4f54da8aedd11c124d8d53ebe727b8203a
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 8f7443494a]
This reverts commit f94be2078c.
Change-Id: I954326d9a291280de7c9b7ef49a8cce0cd9dd9ad
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: a809a5bf32]
This reverts commit 559bb50c6a.
Change-Id: Ibd32ac36bcb9ee418957d25b240ef9dd9172fbd3
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 91b0fce535]
This reverts commit 14d821391e.
Change-Id: Idf636655ac01f569d8e92e9c0e100e5db7d1cb61
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 445fcd803b]
This reverts commit 8f735fb9d5.
Reason for revert: aqlprofile has merged
Change-Id: Iee0fbe8ac1e475123273b328949f68f10490d3ca
[ROCm/ROCR-Runtime commit: 268bb1ae84]
g_props is intialized in topology_take_snapshot which needs to call
hsaKmtAcquireSystemProperties. In hsaKmtOpenKFD, it doesn't call hsaKmtAcquireSystemProperties.
So it needs to change parameter of topology_is_svm_needed from node_id
to EngineId to avoid Segmentation fault on gfx902.
Signed-off-by: changzhu <Changfeng.Zhu@amd.com>
Change-Id: Iba8d20a9142510c70927454d26bdcaf579ad5574
[ROCm/ROCR-Runtime commit: 978028d3f6]
Use the database provided by libdrm-amdgpu,as this is maintained by AMD
Signed-off-by: Mike Li <Tianxinmike.Li@amd.com>
Change-Id: I319a46d833cb23173a77fdff1deae69ea58588b8
[ROCm/ROCR-Runtime commit: cb1c84d99b]
hsaKmtQueryPointerInfo return vm_obj flags for all below registered
memory types other than hsaKmtAllocMemory, and set the CoarseGrain flag
correctly for:
Graphics: always coarse grain.
Shared: hsaKmtShareMemory pass mflags with export handle to KFD to store
in KFD objs, hsaKmtRegisterSharedHandle get mflags from KFD with import
handle.
Userptr: it is already coarse-grain by default, or based on mflags
provided in hsaKmtRegisterMemoryWithFlags call.
Change-Id: Idc23e8b0cf599b02580737639da2f9ef4ccd0c0d
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: fa7b4a6268]
Query pointer info returns KFD_IOC_ALLOC_MEM_FLAGS_* flags, it should
return HsaMemFlags, fix it by renaming vm_obj->flags to mflags and
always saving HsaMemFlags.
Use consistent function parameter and variable name to avoid confusion:
mflags for HsaMemFlags and ioc_flags for KFD_IOC_ALLOC_MEM_FLAGS_*
flags.
AMDKFD_IOC_GET_DMABUF_INFO return ioc_flags, translate it to mflags
using new helper fmm_translate_ioc_to_hsa_flags.
Change-Id: If9e117c507139c0166abb1ab0df8c233ef7e48a1
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: 2c796e62be]
__fmm_release is sometimes called with the aperture lock, and sometimes
without. Consistently call it with the aperture lock held and remove the
lock/unlock calls from this function.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I80dddc64cc0703e5eed8e9f1eb65b75a2c7ae2eb
[ROCm/ROCR-Runtime commit: 5fac7dcc3b]
Unlock mutex if MMIO mapping fails. This happens on all GFXv8 GPUs.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Change-Id: I1dee1cbddefd9185c24ea79377f49f8ae2c5ff57
[ROCm/ROCR-Runtime commit: 19536080a8]
Aperture locking is too fine-grained, it has race between find userptr
and allocate userptr object.
Change _fmm_allocate_device and fmm_allocate_memory_object to not take
the aperture lock, the callers take it, this implements an atomic find
userptr or allocate a new one.
Change-Id: I6773404e22c1f4382a211c5a9817df23c5534a2a
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
[ROCm/ROCR-Runtime commit: c4d5ee28f0]
it is to optimize memory allocation latency, which
changes alignment from 2MB to 1GB.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Change-Id: I7818e9f13b17e2c0992e75b17f978dc03a018a57
[ROCm/ROCR-Runtime commit: 973b35bc06]
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]
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]
Xiaogang Chen
Bing Ma
xinhui pan
Graham Sider
Alex Sierra
Felix Kuehling
Eric Huang
Philip Yang
David Yat Sin
Timothy Pearson
Kent Russell
changzhu
Mike Li
Oak Zeng