Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Kalyan Thota <quic_kalyant@quicinc.com>
To:     y@qualcomm.com, dri-devel@lists.freedesktop.org,
        linux-arm-msm@vger.kernel.org, freedreno@lists.freedesktop.org,
        devicetree@vger.kernel.org
Cc:     Kalyan Thota <quic_kalyant@quicinc.com>,
        linux-kernel@vger.kernel.org, robdclark@gmail.com,
        dianders@chromium.org, swboyd@chromium.org,
        quic_vpolimer@quicinc.com, dmitry.baryshkov@linaro.org,
        quic_abhinavk@quicinc.com
Subject: [v3] drm/msm/disp/dpu1: add support for dspp sub block flush in sc7280
Date:   Wed,  7 Sep 2022 23:16:10 -0700
Message-Id: <1662617770-2795-1-git-send-email-quic_kalyant@quicinc.com>
In-Reply-To: <y>
References: <y>
Precedence: bulk

Flush mechanism for DSPP blocks has changed in sc7280 family, it
allows individual sub blocks to be flushed in coordination with
master flush control.

Representation: master_flush && (PCC_flush | IGC_flush .. etc )

This change adds necessary support for the above design.

Changes in v1:
- Few nits (Doug, Dmitry)
- Restrict sub-block flush programming to dpu_hw_ctl file (Dmitry)

Changes in v2:
- Move the address offset to flush macro (Dmitry)
- Seperate ops for the sub block flush (Dmitry)
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c       |  2 +-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c |  5 +++-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h |  2 ++
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c     | 35 ++++++++++++++++++++++++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h     | 10 ++++++--
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h    |  7 ++++++
 6 files changed, 55 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index 601d687..ab38a52 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -766,7 +766,7 @@ static void _dpu_crtc_setup_cp_blocks(struct drm_crtc *crtc)
 
 		/* stage config flush mask */
 		ctl->ops.update_pending_flush_dspp(ctl,
-			mixer[i].hw_dspp->idx);
+			mixer[i].hw_dspp->idx, DPU_DSPP_SUB_PCC);
 	}
 }
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index 27f029f..0eecb2f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -65,7 +65,10 @@
 	(PINGPONG_SDM845_MASK | BIT(DPU_PINGPONG_TE2))
 
 #define CTL_SC7280_MASK \
-	(BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_FETCH_ACTIVE) | BIT(DPU_CTL_VM_CFG))
+	(BIT(DPU_CTL_ACTIVE_CFG) | \
+	 BIT(DPU_CTL_FETCH_ACTIVE) | \
+	 BIT(DPU_CTL_VM_CFG) | \
+	 BIT(DPU_CTL_DSPP_SUB_BLOCK_FLUSH))
 
 #define MERGE_3D_SM8150_MASK (0)
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index 38aa38a..6a0b784 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -191,6 +191,7 @@ enum {
  * @DPU_CTL_SPLIT_DISPLAY:	CTL supports video mode split display
  * @DPU_CTL_FETCH_ACTIVE:	Active CTL for fetch HW (SSPPs)
  * @DPU_CTL_VM_CFG:		CTL config to support multiple VMs
+ * @DPU_CTL_DSPP_BLOCK_FLUSH: CTL config to support dspp sub-block flush
  * @DPU_CTL_MAX
  */
 enum {
@@ -198,6 +199,7 @@ enum {
 	DPU_CTL_ACTIVE_CFG,
 	DPU_CTL_FETCH_ACTIVE,
 	DPU_CTL_VM_CFG,
+	DPU_CTL_DSPP_SUB_BLOCK_FLUSH,
 	DPU_CTL_MAX
 };
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
index a35ecb6..31c8c44 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c
@@ -33,6 +33,7 @@
 #define   CTL_INTF_FLUSH                0x110
 #define   CTL_INTF_MASTER               0x134
 #define   CTL_FETCH_PIPE_ACTIVE         0x0FC
+#define   CTL_DSPP_n_FLUSH(n)		((0x13C) + ((n - 1) * 4))
 
 #define CTL_MIXER_BORDER_OUT            BIT(24)
 #define CTL_FLUSH_MASK_CTL              BIT(17)
@@ -287,8 +288,9 @@ static void dpu_hw_ctl_update_pending_flush_merge_3d_v1(struct dpu_hw_ctl *ctx,
 }
 
 static void dpu_hw_ctl_update_pending_flush_dspp(struct dpu_hw_ctl *ctx,
-	enum dpu_dspp dspp)
+	enum dpu_dspp dspp, enum dpu_dspp_sub_blk dspp_sub_blk)
 {
+
 	switch (dspp) {
 	case DSPP_0:
 		ctx->pending_flush_mask |= BIT(13);
@@ -307,6 +309,31 @@ static void dpu_hw_ctl_update_pending_flush_dspp(struct dpu_hw_ctl *ctx,
 	}
 }
 
+static void dpu_hw_ctl_update_pending_flush_dspp_subblocks(
+	struct dpu_hw_ctl *ctx,	enum dpu_dspp dspp, enum dpu_dspp_sub_blk dspp_sub_blk)
+{
+	uint32_t flushbits = 0, active;
+
+	switch (dspp_sub_blk) {
+	case DPU_DSPP_SUB_IGC:
+		flushbits = BIT(2);
+		break;
+	case DPU_DSPP_SUB_PCC:
+		flushbits = BIT(4);
+		break;
+	case DPU_DSPP_SUB_GC:
+		flushbits = BIT(5);
+		break;
+	default:
+		return;
+	}
+
+	active = DPU_REG_READ(&ctx->hw, CTL_DSPP_n_FLUSH(dspp));
+	DPU_REG_WRITE(&ctx->hw, CTL_DSPP_n_FLUSH(dspp), active | flushbits);
+
+	ctx->pending_flush_mask |= BIT(29);
+}
+
 static u32 dpu_hw_ctl_poll_reset_status(struct dpu_hw_ctl *ctx, u32 timeout_us)
 {
 	struct dpu_hw_blk_reg_map *c = &ctx->hw;
@@ -675,7 +702,11 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops,
 	ops->setup_blendstage = dpu_hw_ctl_setup_blendstage;
 	ops->update_pending_flush_sspp = dpu_hw_ctl_update_pending_flush_sspp;
 	ops->update_pending_flush_mixer = dpu_hw_ctl_update_pending_flush_mixer;
-	ops->update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp;
+	if (cap & BIT(DPU_CTL_DSPP_SUB_BLOCK_FLUSH))
+		ops->update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp_subblocks;
+	else
+		ops->update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp;
+
 	if (cap & BIT(DPU_CTL_FETCH_ACTIVE))
 		ops->set_active_pipes = dpu_hw_ctl_set_fetch_pipe_active;
 };
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
index 96c012e..227f1bd 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h
@@ -149,12 +149,18 @@ struct dpu_hw_ctl_ops {
 
 	/**
 	 * OR in the given flushbits to the cached pending_flush_mask
-	 * No effect on hardware
+	 *
+	 * If the hardware supports dspp sub block flush, then sub-block
+	 * flushes are written to the hardware and main dspp flush will
+	 * be cached in the pending_flush_mask.
+	 *
 	 * @ctx       : ctl path ctx pointer
 	 * @blk       : DSPP block index
+	 * @dspp_sub_blk : DSPP sub-block index
 	 */
 	void (*update_pending_flush_dspp)(struct dpu_hw_ctl *ctx,
-		enum dpu_dspp blk);
+		enum dpu_dspp blk,  enum dpu_dspp_sub_blk dspp_sub_blk);
+
 	/**
 	 * Write the value of the pending_flush_mask to hardware
 	 * @ctx       : ctl path ctx pointer
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
index d3b0ed0..c113d52 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h
@@ -167,6 +167,13 @@ enum dpu_dspp {
 	DSPP_MAX
 };
 
+enum dpu_dspp_sub_blk{
+	DPU_DSPP_SUB_PCC = 1,
+	DPU_DSPP_SUB_IGC,
+	DPU_DSPP_SUB_GC,
+	DPU_DSPP_SUB_MAX
+};
+
 enum dpu_ctl {
 	CTL_0 = 1,
 	CTL_1,
-- 
2.7.4