Commit 6ec04e38 authored by Jesse.zhang@amd.com's avatar Jesse.zhang@amd.com Committed by Alex Deucher
Browse files

drm/amdgpu/sdma_v4_4_2: update VM flush implementation for SDMA



This commit updates the VM flush implementation for the SDMA engine.

- Added a new function `sdma_v4_4_2_get_invalidate_req` to construct the VM_INVALIDATE_ENG0_REQ
  register value for the specified VMID and flush type. This function ensures that all relevant
  page table cache levels (L1 PTEs, L2 PTEs, and L2 PDEs) are invalidated.

- Modified the `sdma_v4_4_2_ring_emit_vm_flush` function to use the new `sdma_v4_4_2_get_invalidate_req`
  function. The updated function emits the necessary register writes and waits to perform a VM flush
  for the specified VMID. It updates the PTB address registers and issues a VM invalidation request
  using the specified VM invalidation engine.

- Included the necessary header file `gc/gc_9_0_sh_mask.h` to provide access to the required register
  definitions.

v2: vm flush by the vm inalidation packet (Lijo)
v3: code stle and define thh macro for the vm invalidation packet (Christian)
v4: Format definition sdma vm invalidate packet (Lijo)

Suggested-by: default avatarLijo Lazar <lijo.lazar@amd.com>
Signed-off-by: default avatarJesse Zhang <jesse.zhang@amd.com>
Reviewed-by: default avatarLijo Lazar <lijo.lazar@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent b09cdeb4
Loading
Loading
Loading
Loading
+63 −14
Original line number Diff line number Diff line
@@ -31,6 +31,7 @@
#include "amdgpu_ucode.h"
#include "amdgpu_trace.h"
#include "amdgpu_reset.h"
#include "gc/gc_9_0_sh_mask.h"

#include "sdma/sdma_4_4_2_offset.h"
#include "sdma/sdma_4_4_2_sh_mask.h"
@@ -1290,21 +1291,71 @@ static void sdma_v4_4_2_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
			       seq, 0xffffffff, 4);
}


/**
 * sdma_v4_4_2_ring_emit_vm_flush - vm flush using sDMA
/*
 * sdma_v4_4_2_get_invalidate_req - Construct the VM_INVALIDATE_ENG0_REQ register value
 * @vmid: The VMID to invalidate
 * @flush_type: The type of flush (0 = legacy, 1 = lightweight, 2 = heavyweight)
 *
 * @ring: amdgpu_ring pointer
 * @vmid: vmid number to use
 * @pd_addr: address
 * This function constructs the VM_INVALIDATE_ENG0_REQ register value for the specified VMID
 * and flush type. It ensures that all relevant page table cache levels (L1 PTEs, L2 PTEs, and
 * L2 PDEs) are invalidated.
 */
static uint32_t sdma_v4_4_2_get_invalidate_req(unsigned int vmid,
						uint32_t flush_type)
{
	u32 req = 0;

	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);

	return req;
}

/*
 * sdma_v4_4_2_ring_emit_vm_flush - Emit VM flush commands for SDMA
 * @ring: The SDMA ring
 * @vmid: The VMID to flush
 * @pd_addr: The page directory address
 *
 * Update the page table base and flush the VM TLB
 * using sDMA.
 * This function emits the necessary register writes and waits to perform a VM flush for the
 * specified VMID. It updates the PTB address registers and issues a VM invalidation request
 * using the specified VM invalidation engine.
 */
static void sdma_v4_4_2_ring_emit_vm_flush(struct amdgpu_ring *ring,
					 unsigned vmid, uint64_t pd_addr)
					    unsigned int vmid, uint64_t pd_addr)
{
	amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr);
	struct amdgpu_device *adev = ring->adev;
	uint32_t req = sdma_v4_4_2_get_invalidate_req(vmid, 0);
	unsigned int eng = ring->vm_inv_eng;
	struct amdgpu_vmhub *hub = &adev->vmhub[ring->vm_hub];

	amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_lo32 +
				(hub->ctx_addr_distance * vmid),
				lower_32_bits(pd_addr));

	amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_hi32 +
				(hub->ctx_addr_distance * vmid),
				upper_32_bits(pd_addr));
	/*
	 * Construct and emit the VM invalidation packet
	 */
	amdgpu_ring_write(ring,
		SDMA_PKT_VM_INVALIDATION_HEADER_OP(SDMA_OP_VM_INVALIDATE) |
		SDMA_PKT_VM_INVALIDATION_HEADER_SUB_OP(SDMA_SUBOP_VM_INVALIDATE) |
		SDMA_PKT_VM_INVALIDATION_HEADER_XCC0_ENG_ID(0x1f) |
		SDMA_PKT_VM_INVALIDATION_HEADER_XCC1_ENG_ID(0x1f) |
		SDMA_PKT_VM_INVALIDATION_HEADER_MMHUB_ENG_ID(eng));
	amdgpu_ring_write(ring, SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_INVALIDATEREQ(req));
	amdgpu_ring_write(ring, 0);
	amdgpu_ring_write(ring, SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_INVALIDATEACK(BIT(vmid)));
}

static void sdma_v4_4_2_ring_emit_wreg(struct amdgpu_ring *ring,
@@ -2126,8 +2177,7 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = {
		3 + /* hdp invalidate */
		6 + /* sdma_v4_4_2_ring_emit_pipeline_sync */
		/* sdma_v4_4_2_ring_emit_vm_flush */
		SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 +
		4 + 2 * 3 +
		10 + 10 + 10, /* sdma_v4_4_2_ring_emit_fence x3 for user fence, vm fence */
	.emit_ib_size = 7 + 6, /* sdma_v4_4_2_ring_emit_ib */
	.emit_ib = sdma_v4_4_2_ring_emit_ib,
@@ -2159,8 +2209,7 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = {
		3 + /* hdp invalidate */
		6 + /* sdma_v4_4_2_ring_emit_pipeline_sync */
		/* sdma_v4_4_2_ring_emit_vm_flush */
		SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 +
		4 + 2 * 3 +
		10 + 10 + 10, /* sdma_v4_4_2_ring_emit_fence x3 for user fence, vm fence */
	.emit_ib_size = 7 + 6, /* sdma_v4_4_2_ring_emit_ib */
	.emit_ib = sdma_v4_4_2_ring_emit_ib,
+70 −0
Original line number Diff line number Diff line
@@ -64,6 +64,9 @@
#define HEADER_BARRIER  5
#define SDMA_OP_AQL_COPY  0
#define SDMA_OP_AQL_BARRIER_OR  0
/* vm invalidation is only available for GC9.4.3/GC9.4.4/GC9.5.0 */
#define SDMA_OP_VM_INVALIDATE 8
#define SDMA_SUBOP_VM_INVALIDATE 4

/*define for op field*/
#define SDMA_PKT_HEADER_op_offset 0
@@ -3331,5 +3334,72 @@
#define SDMA_AQL_PKT_BARRIER_OR_COMPLETION_SIGNAL_HI_completion_signal_63_32_shift  0
#define SDMA_AQL_PKT_BARRIER_OR_COMPLETION_SIGNAL_HI_COMPLETION_SIGNAL_63_32(x) (((x) & SDMA_AQL_PKT_BARRIER_OR_COMPLETION_SIGNAL_HI_completion_signal_63_32_mask) << SDMA_AQL_PKT_BARRIER_OR_COMPLETION_SIGNAL_HI_completion_signal_63_32_shift)

/*
** Definitions for SDMA_PKT_VM_INVALIDATION packet
*/

/*define for HEADER word*/
/*define for op field*/
#define SDMA_PKT_VM_INVALIDATION_HEADER_op_offset 0
#define SDMA_PKT_VM_INVALIDATION_HEADER_op_mask   0x000000FF
#define SDMA_PKT_VM_INVALIDATION_HEADER_op_shift  0
#define SDMA_PKT_VM_INVALIDATION_HEADER_OP(x) ((x & SDMA_PKT_VM_INVALIDATION_HEADER_op_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_op_shift)

/*define for sub_op field*/
#define SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_offset 0
#define SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_mask   0x000000FF
#define SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_shift  8
#define SDMA_PKT_VM_INVALIDATION_HEADER_SUB_OP(x) ((x & SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_shift)

/*define for xcc0_eng_id field*/
#define SDMA_PKT_VM_INVALIDATION_HEADER_xcc0_eng_id_offset 0
#define SDMA_PKT_VM_INVALIDATION_HEADER_xcc0_eng_id_mask   0x0000001F
#define SDMA_PKT_VM_INVALIDATION_HEADER_xcc0_eng_id_shift  16
#define SDMA_PKT_VM_INVALIDATION_HEADER_XCC0_ENG_ID(x) ((x & SDMA_PKT_VM_INVALIDATION_HEADER_xcc0_eng_id_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_xcc0_eng_id_shift)

/*define for xcc1_eng_id field*/
#define SDMA_PKT_VM_INVALIDATION_HEADER_xcc1_eng_id_offset 0
#define SDMA_PKT_VM_INVALIDATION_HEADER_xcc1_eng_id_mask   0x0000001F
#define SDMA_PKT_VM_INVALIDATION_HEADER_xcc1_eng_id_shift  21
#define SDMA_PKT_VM_INVALIDATION_HEADER_XCC1_ENG_ID(x) ((x & SDMA_PKT_VM_INVALIDATION_HEADER_xcc1_eng_id_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_xcc1_eng_id_shift)

/*define for mmhub_eng_id field*/
#define SDMA_PKT_VM_INVALIDATION_HEADER_mmhub_eng_id_offset 0
#define SDMA_PKT_VM_INVALIDATION_HEADER_mmhub_eng_id_mask   0x0000001F
#define SDMA_PKT_VM_INVALIDATION_HEADER_mmhub_eng_id_shift  26
#define SDMA_PKT_VM_INVALIDATION_HEADER_MMHUB_ENG_ID(x) ((x & SDMA_PKT_VM_INVALIDATION_HEADER_mmhub_eng_id_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_mmhub_eng_id_shift)

/*define for INVALIDATEREQ word*/
/*define for invalidatereq field*/
#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_offset 1
#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_mask   0xFFFFFFFF
#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_shift  0
#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_INVALIDATEREQ(x) ((x & SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_mask) << SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_shift)

/*define for ADDRESSRANGELO word*/
/*define for addressrangelo field*/
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_offset 2
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_mask   0xFFFFFFFF
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_shift  0
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_ADDRESSRANGELO(x) ((x & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_shift)

/*define for ADDRESSRANGEHI word*/
/*define for invalidateack field*/
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_offset 3
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_mask   0x0000FFFF
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_shift  0
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_INVALIDATEACK(x) ((x & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_shift)

/*define for addressrangehi field*/
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_offset 3
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_mask   0x0000001F
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_shift  16
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_ADDRESSRANGEHI(x) ((x & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_shift)

/*define for reserved field*/
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_offset 3
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_mask   0x000001FF
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_shift  23
#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_RESERVED(x) ((x & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_shift)

#endif /* __SDMA_PKT_OPEN_H_ */