- 09 Jun, 2023 40 commits
-
-
Rajneesh Bhardwaj authored
On GPXIP 9.4.3 APU, in no carveout mode there is no real vram heap and could be emulated by the driver over the interleaved NUMA system memory and the APU could also be in the carveout mode during early development stage or otherwise for debugging purpose so introduce a new member in amdgpu_gmc to figure out whether the APU is in the native mode as per the production configuration. AMD_IS_APU cannot be used for Accelerated Processing Platform APUs as it might be used in a different context on previous generations or on small APUs. Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Tested-by: Graham Sider <graham.sider@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Philip Yang authored
Output IH cookie node_id and translate it to the corresponding AID id and XCC id, to help debug the GPU page fault. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Shiwu Zhang authored
For driver de-init like rmmod operations those partition specific attributes need to be removed accordingly. Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com> Reviewed-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Shiwu Zhang authored
For gfx_v9_4_3 and beyond, struct kiq has its own mqd_backup pointer rather than using the last pointer from mec struct. Then the kfree operation on the pointer from the mec struct should be removed otherwise it will cause double free on the first kcq's mqd_backup buffer on XCD1. Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
Skip reading runtime db information for PSP 13.0.6. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Assign the vm context register addr per aid instance. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
Programming of this register is taken care by PSP. Incorrect programming causes CP not to detect its XCC. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reported-by: Alexander Turek <Alexander.Turek@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Rajneesh Bhardwaj authored
- Add helpers to detect the current GPU memory partition. - Add current memory partition mode sysfs node. Tested-by: Ori Messinger <Ori.Messinger@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Hawking Zhang authored
Add callbacks for SMUIO 13.0.3 Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Rajneesh Bhardwaj authored
Add smuio v13_0_3 callbacks for SMUIO. Tested-by: Ori Messinger <Ori.Messinger@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Hawking Zhang authored
Add smuio v13_0_3 register offset and shift masks header files v2: update headers (Alex) Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Hawking Zhang authored
gfx v9_4_3 only support compute. render backend doesn't need to be involved in any compute shader execution. Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Tom St Denis authored
This patch updates the 'regs2' interface for MMIO registers to add a new IOCTL command for a 'v2' state data that includes the XCC ID. This patch then updates amdgpu_gfx_select_se_sh() and amdgpu_gfx_select_me_pipe_q() (and the implementations in the gfx drivers) to support an additional parameter. This patch then creates a new debugfs interface "gprwave" which is a merge of shader GPR and wave status access. This new inteface uses an IOCTL to select banks as well as XCC identity. (v2) Fix missing xcc_id in wave_ind function (v3) Fix pm runtime calls and mutex locking (v4) Fix bad label Signed-off-by: Tom St Denis <tom.stdenis@amd.com> Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
Add programming of SDMA golden settings for v4.4.2 Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Hawking Zhang authored
it is used for user space driver to identify gfx_v9_4_3 chip Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
IH follows a different identification scheme for its clients. Get the right mapping of xcc instance from IH node id. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Use WREG32_SOC15_EXT to write registers with address larger than 32bit. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
SMN address which is larger than 32bit has different indications through bit[34:32] on different AIDs. v2: put smn addressing of different AIDs into asic specific place v3: change to ext_id/ext_offset naming Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Truncate the 64bit base address from ip discovery and only store lower 32bit ip base in reg_offset[]. Bits > 32 follows ASIC specific format, thus just discard them and handle it within specific ASIC. By this way reg_offset[] and related helpers can stay unchanged. v2: make comments more generic Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
version 4 supports 64bit ip base address Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Use WREG_SOC15x() instead of WREG32(SOC15_REG_OFFSET()) Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Number of instances is extended. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
New ip_discovery binary size is increased. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
version 4 supports 64bit ip base address Signed-off-by: Le Ma <le.ma@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
New doorbell index assignment is used by aqua_vanjaram. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
For aqua vanjaram, add mapping for logical to physical instances. v2: Register accesses on bare metal should be based on physical instance. Use GET_INST() to get physical instance. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
Add a mask of SDMA instances available for use. On certain ASIC configs, not all SDMA instances are available for software use. v2: Change sdma mask type to uint32_t (Le) Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
Add XCC logical to physical instance map for aqua vanjaram v2: Keep look up table only for required IPs, for others return default mapping (Felix). Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <Le.Ma@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Four basic reasons as below to do the change: 1. number of ring expand a lot on aqua_vanjaram, and adjustment on old assignment cannot make each ring in a continuous doorbell space. 2. the SDMA doorbell index should not exceed 0x1FF on aqua_vanjaram due to regDOORBELLx_CTRL_ENTRY.BIF_DOORBELLx_RANGE_OFFSET_ENTRY field width. 3. re-design the doorbell assignment and unify the calculation as "start + ring/inst id" will make the code much concise. 4. only defining the START/END makes the table look simple v2: (Lijo) 1. replace name 2. use num_inst_per_aid/sdma_doorbell_range instead of hardcoding Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
In GC v9.4.3 there are multiple XCCs. It's required to use physical instance number to get the right register offset. Use GET_INST API for that. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
On initialization set the partition mode correctly to SPX (default) or any other user specified partition mode. Use switch_compute_partition API so that all settings are initialized correctly. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Mukul Joshi authored
Update interrupt handling in CPX mode for GFX9.4.3 by using the VMID space instead of SDMA client id to determine if an interrupt should be processed by a KFD node. This is especially needed for handling retry faults from MMHUB. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Mukul Joshi authored
Fix the if condition which causes dynamic repartitioning to fail when trying to switch to DPX mode. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Amber Lin <Amber.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Mukul Joshi authored
For GFX 9.4.3, use the logical to physical mapping table, to get the correct XCD instance when accessing registers on bare metal. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Amber Lin <Amber.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Amber Lin authored
GFX_9_4_3 supports multi-XCDs and multi-AIDs in one GPU device. SWS needs to program IH_VMID_x_LUT with specified XCC instance and corresponded AID instance. Signed-off-by: Amber Lin <Amber.Lin@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
It's not required for compute pipeline and will cause soft lockup on emulation due to long-time writing. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Le Ma authored
Use s2a entry 5/6 registers to decode sdma doorbell trans on different AIDs, which aligns the entry table in SHUB spec, and leave entry 4 dedicated for VCN doorbell to avoid conflict. Signed-off-by: Le Ma <le.ma@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Mukul Joshi authored
On GFX 9.4.3, there can be multiple KFD nodes. As a result, SMI events for SVM, queue evict/restore should be raised for each node independently. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Lijo Lazar authored
Program partition status register to reflect the current partition mode. Partition capability register is for capability and is a one-time setting. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-
Alex Sierra authored
This work is required for GC 9.4.3, previous to support memory partitions per node at SVM. When multiple partition is configured, every BO should be allocated inside one specific partition which corresponds to the current amdgpu_device and kfd_node. v2: squash in compilation fix (Alex) v3: squash in fix for pre-gfx 9.4.3 (Alex) v4: squash in best_loc fix (Alex) Signed-off-by: Alex Sierra <alex.sierra@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-