[PATCH 5/6] drm/amdgpu:use FRAME_CNTL for new GFX ucode

[Monk Liu]

VI/AI affected:

CP/HW team requires KMD insert FRAME_CONTROL(end) after
the last IB and before the fence of this DMAframe.

this is to make sure the cache are flushed, and it's a must
change no matter MCBP/SR-IOV or bare-metal case because new
CP hw won't do the cache flush for each IB anymore, it just
leaves it to KMD now.

with this patch, certain MCBP hang issue when rendering
vulkan/chained-ib are resolved.

Change-Id: I34ee7528aa32e704b2850bc6d50774b24c29b840
Signed-off-by: Monk Liu <Monk.Liu at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c   |  3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h |  1 +
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c    | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c    | 11 +++++++++++
 include/uapi/drm/amdgpu_drm.h            |  2 ++
 5 files changed, 28 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 4480e01..df49709 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -201,6 +201,9 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 			!amdgpu_sriov_vf(adev)) /* for SRIOV preemption, Preamble CE ib must be inserted anyway */
 			continue;
 
+		if (i == num_ibs - 1)
+			ib->flags |= AMDGPU_IB_FLAG_LAST;
+
 		amdgpu_ring_emit_ib(ring, ib, job ? job->vm_id : 0,
 				    need_ctx_switch);
 		need_ctx_switch = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 2acaac6..03e88c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -143,6 +143,7 @@ struct amdgpu_ring_funcs {
 	void (*emit_cntxcntl) (struct amdgpu_ring *ring, uint32_t flags);
 	void (*emit_rreg)(struct amdgpu_ring *ring, uint32_t reg);
 	void (*emit_wreg)(struct amdgpu_ring *ring, uint32_t reg, uint32_t val);
+	void (*emit_tmz)(struct amdgpu_ring *ring, bool start);
 };
 
 struct amdgpu_ring {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index eed5745..f4aeba6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -6422,6 +6422,10 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 			  (ib->gpu_addr & 0xFFFFFFFC));
 	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
 	amdgpu_ring_write(ring, control);
+
+	/* insert FRAME_CONTROL (end) after the last IB */
+	if (ib->flags & AMDGPU_IB_FLAG_LAST && ring->funcs->emit_tmz)
+		ring->funcs->emit_tmz(ring, false);
 }
 
 static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
@@ -6651,6 +6655,12 @@ static void gfx_v8_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigne
 		ring->ring[offset] = (ring->ring_size >> 2) - offset + cur;
 }
 
+static void gfx_v8_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0));
+	amdgpu_ring_write(ring, FRAME_CMD(start ? 0 : 1)); /* frame_end */
+}
+
 
 static void gfx_v8_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg)
 {
@@ -6932,6 +6942,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
 	.emit_cntxcntl = gfx_v8_ring_emit_cntxcntl,
 	.init_cond_exec = gfx_v8_0_ring_emit_init_cond_exec,
 	.patch_cond_exec = gfx_v8_0_ring_emit_patch_cond_exec,
+	.emit_tmz = gfx_v8_0_ring_emit_tmz,
 };
 
 static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index f192297..745e2f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2899,6 +2899,10 @@ BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
 		lower_32_bits(ib->gpu_addr));
 	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
 	amdgpu_ring_write(ring, control);
+
+	/* insert FRAME_CONTROL (end) after the last IB */
+	if (ib->flags & AMDGPU_IB_FLAG_LAST && ring->funcs->emit_tmz)
+		ring->funcs->emit_tmz(ring, false);
 }
 
 #define	INDIRECT_BUFFER_VALID                   (1 << 23)
@@ -3156,6 +3160,12 @@ static void gfx_v9_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigne
 		ring->ring[offset] = (ring->ring_size>>2) - offset + cur;
 }
 
+static void gfx_v9_0_ring_emit_tmz(struct amdgpu_ring *ring, bool start)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0));
+	amdgpu_ring_write(ring, FRAME_CMD(start ? 0 : 1)); /* frame_end */
+}
+
 static void gfx_v9_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -3490,6 +3500,7 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = {
 	.emit_cntxcntl = gfx_v9_ring_emit_cntxcntl,
 	.init_cond_exec = gfx_v9_0_ring_emit_init_cond_exec,
 	.patch_cond_exec = gfx_v9_0_ring_emit_patch_cond_exec,
+	.emit_tmz = gfx_v9_0_ring_emit_tmz,
 };
 
 static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_compute = {
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 20ea895..e979285 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -478,6 +478,8 @@ union drm_amdgpu_cs {
 /* Preempt flag, IB should set Pre_enb bit if PREEMPT flag detected */
 #define AMDGPU_IB_FLAG_PREEMPT (1<<2)
 
+#define AMDGPU_IB_FLAG_LAST (1<<3)
+
 struct drm_amdgpu_cs_chunk_ib {
 	__u32 _pad;
 	/** AMDGPU_IB_FLAG_* */
-- 
2.7.4