Commit Graph

38 Commits

Author SHA1 Message Date
Harish Kasiviswanathan e7e1361c3d Use new ioctl for getting process apertures
Change-Id: I73678744ad73942edec442ad9c6d38637f7e1235
2016-01-12 12:09:25 -05:00
Felix Kuehling 063ad3ad9e Implement hsaKmtRegisterMemoryToNodes
Fix hsaKmtRegisterMemory to be a no-op for now and move the multi-GPU
implementation to hsaKmtRegisterMemoryToNodes. Make GPU memory mappings
of host memory visible to all GPUs by default. Device memory is still
visible to the allocating GPU only by default (but can be overridden
with hsaKmtRegisterMemoryToNodes for experimenting with P2P).

Change-Id: I73408afbe3b10c8dad2ab3a780f58413249692e6
2016-01-08 16:00:23 -05:00
Ben Goz ea0f9d2a0b Adding support for mGPU
Change-Id: I5ed184e6a58b38d9dde48867f14513d161cf41a9
Signed-off-by: Ben Goz <ben.goz@amd.com>
2016-01-04 15:35:15 +02:00
Ben Goz e61500c46e Adding support for new AQL Queue Memory allocation
Change-Id: If84fc4b961627dbdd0b77b1c509a3c9a4c709b9f
Signed-off-by: Ben Goz <ben.goz@amd.com>
2015-10-22 13:13:54 +03:00
Felix Kuehling b94ae66c62 Update HsaMemFlags.ui32.CoarseGrain comment
As advised by Paul Blinzer

Change-Id: Icabf4acd94866ddbbe53faf48a71e1113f0c76b6
2015-10-05 16:48:50 -04:00
Felix Kuehling f2f45cc0e4 Add CoarseGrain memory flag
Change-Id: If8ac0339ae8c809c6e6a4f56592a4061d110ea94
2015-09-30 18:07:04 -04:00
shaoyunl 2d63ee7b8f Initiali support for CWSR on thunk
1. Add IOCTL defines to set trap handler
2. Add control stack size information on create queue argument.
3. Increase the total save&restore area size for carrizo to include the control stack size.

Signed-off-by: Shaoyun Liu <Shaoyun.liu@amd.com>

Change-Id: Iccf15e073b7db2519e96e7f7b46a89d57ab9a4df
2015-09-25 15:12:25 -04:00
Amber Lin 082f8314c4 Sync up HSA_ENGINE_ID type with Windows/Perforce
HSA_ENGINE_ID in Perforce added ui32 to the typedef while in Git it doesn't.
This causes conflicts to RT applications. Decision being made is to change Git
to match Perforce.

Change-Id: I7e9c6437b023bb23ec9578737f8534e9453589b9
2015-09-24 00:10:52 -04:00
Ben Goz 692e004047 Adding new memory allocation IOCTL
Change-Id: I0eb1924811a2e1e436296ebe632d8f112a61637d
Signed-off-by: Ben Goz <ben.goz@amd.com>
2015-09-21 13:58:32 -04:00
Ben Goz fb8378a18b Support gfx802 dGPU
Signed-off-by: Ben Goz <ben.goz@amd.com>
2015-08-30 14:13:53 +03:00
Amber Lin a3925a3a19 Enable version info via thunk interface
- Replace HSAuint32 with HSA_ENGINE_ID for EngineId type so it explicitely
  presents version information for ucode and GfxIP
- Created a GfxIP lookup table to pass the version information. This lookup
  searches for matching device ID.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Acked-by: John Bridgman <John.Bridgman@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
2015-07-31 14:56:33 -04:00
Flora Cui fc4e07daa3 Add interface to set CU mask
Signed-off-by: Flora Cui <flora.cui@amd.com>
Acked-by: Ben Goz <ben.goz@amd.com>
2015-07-23 15:44:01 +08:00
Moses Reuben 29c083f695 adding support for scratch memory
Signed-off-by: Moses Reuben <moses.reuben@amd.com>
2015-07-21 16:43:23 +03:00
Oded Gabbay 2e76017278 increase event limit to provide 4K events
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2015-05-18 11:01:42 +03:00
Oded Gabbay a70a98b30b Increase limit of signal events to 4096
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
Reviewed-by: Ben Goz<ben.goz@amd.com>
2015-05-03 13:58:10 +03:00
Oded Gabbay eb2d3cfcdf Add missing DoorbellType field to HSA_CAPABILITY
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2015-05-02 12:10:04 +03:00
Oded Gabbay 4c4df38035 Revert "Add execution property in register memory for gfx801."
This reverts commit 5ed05c99b3.
2015-04-28 17:50:00 +03:00
Xihan Zhang 5ed05c99b3 Add execution property in register memory for gfx801.
Signed-off-by: Xihan Zhang <xihan.zhang@amd.com>
2015-04-10 22:26:44 +08:00
Ben Goz b27de08442 Adding support for mmap events page from Thunk
Signed-off-by: Ben Goz <ben.goz@amd.com>
2015-03-24 17:55:54 +02:00
Oded Gabbay e62d46ba1c Don't alloc/free VRAM in map/unmap paths
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2015-02-25 22:49:11 +02:00
Oded Gabbay 5c0f8660b6 Revert "Workaround for non cached memory issue on gfx801 A0"
This reverts commit 402ff3f881.

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
Acked-by: Ben Goz <ben.goz@amd.com>

Conflicts:
	include/linux/kfd_ioctl.h
	src/queues.c
2015-02-16 17:22:45 +02:00
Yair Shachar 89188a6d6c Modify HW dbg thunk,to adapt to new KFD-thunk interface
Signed-off-by: Yair Shachar <yair.shachar@amd.com>
2015-02-08 21:44:47 +02:00
Oded Gabbay f0a7fcccc9 Match as much as possible hsakmt.h and hsakmttypes.h to p4
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2015-02-03 11:13:13 +02:00
Ben Goz 25441796bd Supporting new thunk spec - adding relevant memflags
Signed-off-by: Ben Goz <ben.goz@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>
2015-02-01 14:33:33 +02:00
Oded Gabbay 8cc0b16293 match kfd_ioctl.h to kernel repo
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2015-01-19 17:18:47 +02:00
Alexey Skidanov d7196d4122 IOCTL: Adding memory exception data
Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>
2014-11-27 16:21:53 +02:00
Alexey Skidanov 36ccd7f867 Adding memory event type and definition of memory exception data
Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>
2014-11-27 15:54:21 +02:00
Ben Goz 402ff3f881 Workaround for non cached memory issue on gfx801 A0
Signed-off-by: Ben Goz <ben.goz@amd.com>

Adjust to new ioctl format

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-07 14:51:53 +02:00
Ben Goz bd7e10b0ec Adding support in allocating executable memory
Signed-off-by: Ben Goz <ben.goz@amd.com>
2014-12-07 14:36:21 +02:00
Oded Gabbay 7972bb00c5 kfd_ioctl: interop
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:02:34 +02:00
Oded Gabbay d995344ae7 Add new definitions of HSA functions and types
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-10-20 08:48:28 +03:00
Oded Gabbay d14eb7aa00 kfd_ioctl: local memory
Add two new IOCTLs

Remove old IOCTLs

Signed-off-by: Alexey Skidanov <Alexey.Skidanov@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:02:27 +02:00
Oded Gabbay f7c3c3bd1b kfd_ioctl: debug
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:02:12 +02:00
Oded Gabbay 03604633ca kfd_ioctl: events
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:01:26 +02:00
Oded Gabbay 0a8bc868b5 set thunk-kfd version to match stg
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-08 10:04:31 +02:00
Oded Gabbay 9f647b07ff Change names of IOCTLS
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-12-29 13:34:32 +02:00
Oded Gabbay 813af2b89a Add definitions of HSA functions and types
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-07-29 10:55:52 +03:00
Oded Gabbay b0b875964f Add IOCTL set definitions of amdkfd
- KFD_IOC_GET_VERSION:
	Retrieves the interface version of amdkfd

- KFD_IOC_CREATE_QUEUE:
	Creates a usermode queue that runs on a specific GPU device

- KFD_IOC_DESTROY_QUEUE:
	Destroys an existing usermode queue

- KFD_IOC_SET_MEMORY_POLICY:
	Sets the memory policy of the default and alternate aperture of the
	calling process

- KFD_IOC_GET_CLOCK_COUNTERS:
	Retrieves counters (timestamps) of CPU and GPU

- KFD_IOC_GET_PROCESS_APERTURES:
	Retrieves information about process apertures that were initialized
	during the open() call of the amdkfd device

- KFD_IOC_UPDATE_QUEUE:
	Updates configuration of an existing usermode queue

Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
2014-07-29 10:52:32 +03:00