drm/amdgpu: Add a max ibs per submission limit.

And ensure each ring supports that many submissions. This makes
sure that we don't get surprises after the submission has been
scheduled where the ring allocation actually gets rejected.

My calculations on the existing limits:
COMPUTE v10: 128
COMPUTE v11: 128
COMPUTE v6: 157
COMPUTE v7: 133
COMPUTE v8: 130
COMPUTE v9: 125
GFX v10: 208
GFX v11: 213
GFX v6: 154 (doubling this in the previous patch)
GFX v7: 226
GFX v8: 213
GFX v9: 208
GFX v9 (SW): 208
SDMA CIK: 87
SDMA SI: 97
SDMA v2.4: 74
SDMA v3.0: 74
SDMA v4.0: 72
SDMA v5.0: 51
SDMA v6.0: 52
UVD ENC v6.0: 98
UVD ENC v7.0: 92
UVD v3.1: 124
UVD v4.2: 124
UVD v5.0: 83
UVD v6.0  (VM): 55
UVD v7.0: 51
VCE v2.0: 126
VCE v3.0 (VM): 98
VCE v4.0: 93
VCN DEC v1.0: 49
VCN DEC v2.0: 51
VCN DEC v3.0: 51
VCN ENC v1.0: 58
VCN ENC v2.0: 93
VCN ENC v3.0: 93
VCN ENC v4.0: 93
VCN JPEG v1.0: 17
VCN JPEG v2.0: 16
VCN JPEG v2.5: 17
VCN JPEG v3.0: 17
VCN JPEG v4.0: 17

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2498Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c

@@ -112,6 +112,9 @@ static int amdgpu_cs_p1_ib(struct amdgpu_cs_parser *p,
 	if (r < 0)
 		return r;
 
+	if (num_ibs[r] >= amdgpu_ring_max_ibs(chunk_ib->ip_type))
+		return -EINVAL;
+
 	++(num_ibs[r]);
 	p->gang_leader_idx = r;
 	return 0;

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c

@@ -49,6 +49,26 @@
  * them until the pointers are equal again.
  */
 
+/**
+ * amdgpu_ring_max_ibs - Return max IBs that fit in a single submission.
+ *
+ * @type: ring type for which to return the limit.
+ */
+unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type)
+{
+	switch (type) {
+	case AMDGPU_RING_TYPE_GFX:
+		/* Need to keep at least 192 on GFX7+ for old radv. */
+		return 192;
+	case AMDGPU_RING_TYPE_COMPUTE:
+		return 125;
+	case AMDGPU_RING_TYPE_VCN_JPEG:
+		return 16;
+	default:
+		return 49;
+	}
+}
+
 /**
  * amdgpu_ring_alloc - allocate space on the ring buffer
  *
@@ -182,6 +202,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 	int sched_hw_submission = amdgpu_sched_hw_submission;
 	u32 *num_sched;
 	u32 hw_ip;
+	unsigned int max_ibs_dw;
 
 	/* Set the hw submission limit higher for KIQ because
 	 * it's used for a number of gfx/compute tasks by both
@@ -290,6 +311,14 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		return r;
 	}
 
+	max_ibs_dw = ring->funcs->emit_frame_size +
+		     amdgpu_ring_max_ibs(ring->funcs->type) * ring->funcs->emit_ib_size;
+	max_ibs_dw = (max_ibs_dw + ring->funcs->align_mask) & ~ring->funcs->align_mask;
+
+	if (WARN_ON(max_ibs_dw > max_dw)) {
+		max_dw = max_ibs_dw;
+	}
+
 	ring->ring_size = roundup_pow_of_two(max_dw * 4 * sched_hw_submission);
 
 	ring->buf_mask = (ring->ring_size / 4) - 1;

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h

@@ -319,6 +319,7 @@ struct amdgpu_ring {
 #define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o))
 #define amdgpu_ring_preempt_ib(r) (r)->funcs->preempt_ib(r)
 
+unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type);
 int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw);
 void amdgpu_ring_ib_begin(struct amdgpu_ring *ring);
 void amdgpu_ring_ib_end(struct amdgpu_ring *ring);