[PATCH v9 11/42] drm/msm: Stop passing vm to msm_framebuffer

[Rob Clark <robin.clark@oss.qualcomm.com>]

The fb only deals with kms->vm, so make that explicit.  This will start
letting us refcount the # of times the fb is pinned, so we can only
unpin the vma after last user of the fb is done.  Having a single
reference count really only works if there is only a single vm.

Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Tested-by: Antonino Maniscalco <antomani103@gmail.com>
Reviewed-by: Antonino Maniscalco <antomani103@gmail.com>
---
 .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c   | 11 +++-------
 drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c   | 18 +++++++----------
 drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h   |  3 +--
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c     | 20 ++++++-------------
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h     |  2 --
 drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c    | 18 ++++++-----------
 drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c    | 18 ++++++-----------
 drivers/gpu/drm/msm/msm_drv.h                 |  9 +++------
 drivers/gpu/drm/msm/msm_fb.c                  | 15 +++++++-------
 9 files changed, 39 insertions(+), 75 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index 32e208ee946d..9a54da1c9e3c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -566,7 +566,6 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc
 		struct drm_writeback_job *job)
 {
 	const struct msm_format *format;
-	struct msm_gem_vm *vm;
 	struct dpu_hw_wb_cfg *wb_cfg;
 	int ret;
 	struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc);
@@ -576,13 +575,12 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc
 
 	wb_enc->wb_job = job;
 	wb_enc->wb_conn = job->connector;
-	vm = phys_enc->dpu_kms->base.vm;
 
 	wb_cfg = &wb_enc->wb_cfg;
 
 	memset(wb_cfg, 0, sizeof(struct dpu_hw_wb_cfg));
 
-	ret = msm_framebuffer_prepare(job->fb, vm, false);
+	ret = msm_framebuffer_prepare(job->fb, false);
 	if (ret) {
 		DPU_ERROR("prep fb failed, %d\n", ret);
 		return;
@@ -596,7 +594,7 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc
 		return;
 	}
 
-	dpu_format_populate_addrs(vm, job->fb, &wb_cfg->dest);
+	dpu_format_populate_addrs(job->fb, &wb_cfg->dest);
 
 	wb_cfg->dest.width = job->fb->width;
 	wb_cfg->dest.height = job->fb->height;
@@ -619,14 +617,11 @@ static void dpu_encoder_phys_wb_cleanup_wb_job(struct dpu_encoder_phys *phys_enc
 		struct drm_writeback_job *job)
 {
 	struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc);
-	struct msm_gem_vm *vm;
 
 	if (!job->fb)
 		return;
 
-	vm = phys_enc->dpu_kms->base.vm;
-
-	msm_framebuffer_cleanup(job->fb, vm, false);
+	msm_framebuffer_cleanup(job->fb, false);
 	wb_enc->wb_job = NULL;
 	wb_enc->wb_conn = NULL;
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
index d115b79af771..b0d585c5315c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c
@@ -274,15 +274,14 @@ int dpu_format_populate_plane_sizes(
 	return _dpu_format_populate_plane_sizes_linear(fmt, fb, layout);
 }
 
-static void _dpu_format_populate_addrs_ubwc(struct msm_gem_vm *vm,
-					    struct drm_framebuffer *fb,
+static void _dpu_format_populate_addrs_ubwc(struct drm_framebuffer *fb,
 					    struct dpu_hw_fmt_layout *layout)
 {
 	const struct msm_format *fmt;
 	uint32_t base_addr = 0;
 	bool meta;
 
-	base_addr = msm_framebuffer_iova(fb, vm, 0);
+	base_addr = msm_framebuffer_iova(fb, 0);
 
 	fmt = msm_framebuffer_format(fb);
 	meta = MSM_FORMAT_IS_UBWC(fmt);
@@ -355,26 +354,23 @@ static void _dpu_format_populate_addrs_ubwc(struct msm_gem_vm *vm,
 	}
 }
 
-static void _dpu_format_populate_addrs_linear(struct msm_gem_vm *vm,
-					      struct drm_framebuffer *fb,
+static void _dpu_format_populate_addrs_linear(struct drm_framebuffer *fb,
 					      struct dpu_hw_fmt_layout *layout)
 {
 	unsigned int i;
 
 	/* Populate addresses for simple formats here */
 	for (i = 0; i < layout->num_planes; ++i)
-		layout->plane_addr[i] = msm_framebuffer_iova(fb, vm, i);
+		layout->plane_addr[i] = msm_framebuffer_iova(fb, i);
 	}
 
 /**
  * dpu_format_populate_addrs - populate buffer addresses based on
  *                     mmu, fb, and format found in the fb
- * @vm:                address space pointer
  * @fb:                framebuffer pointer
  * @layout:            format layout structure to populate
  */
-void dpu_format_populate_addrs(struct msm_gem_vm *vm,
-			       struct drm_framebuffer *fb,
+void dpu_format_populate_addrs(struct drm_framebuffer *fb,
 			       struct dpu_hw_fmt_layout *layout)
 {
 	const struct msm_format *fmt;
@@ -384,7 +380,7 @@ void dpu_format_populate_addrs(struct msm_gem_vm *vm,
 	/* Populate the addresses given the fb */
 	if (MSM_FORMAT_IS_UBWC(fmt) ||
 			MSM_FORMAT_IS_TILE(fmt))
-		_dpu_format_populate_addrs_ubwc(vm, fb, layout);
+		_dpu_format_populate_addrs_ubwc(fb, layout);
 	else
-		_dpu_format_populate_addrs_linear(vm, fb, layout);
+		_dpu_format_populate_addrs_linear(fb, layout);
 }
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h
index 989f3e13c497..dc03f522e616 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h
@@ -31,8 +31,7 @@ static inline bool dpu_find_format(u32 format, const u32 *supported_formats,
 	return false;
 }
 
-void dpu_format_populate_addrs(struct msm_gem_vm *vm,
-			       struct drm_framebuffer *fb,
+void dpu_format_populate_addrs(struct drm_framebuffer *fb,
 			       struct dpu_hw_fmt_layout *layout);
 
 int dpu_format_populate_plane_sizes(
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 6d47f43f52f7..07f0461223c3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -646,7 +646,6 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane,
 	struct drm_framebuffer *fb = new_state->fb;
 	struct dpu_plane *pdpu = to_dpu_plane(plane);
 	struct dpu_plane_state *pstate = to_dpu_plane_state(new_state);
-	struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base);
 	int ret;
 
 	if (!new_state->fb)
@@ -654,9 +653,6 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane,
 
 	DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", fb->base.id);
 
-	/* cache vm */
-	pstate->vm = kms->base.vm;
-
 	/*
 	 * TODO: Need to sort out the msm_framebuffer_prepare() call below so
 	 *       we can use msm_atomic_prepare_fb() instead of doing the
@@ -664,13 +660,10 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane,
 	 */
 	drm_gem_plane_helper_prepare_fb(plane, new_state);
 
-	if (pstate->vm) {
-		ret = msm_framebuffer_prepare(new_state->fb,
-				pstate->vm, pstate->needs_dirtyfb);
-		if (ret) {
-			DPU_ERROR("failed to prepare framebuffer\n");
-			return ret;
-		}
+	ret = msm_framebuffer_prepare(new_state->fb, pstate->needs_dirtyfb);
+	if (ret) {
+		DPU_ERROR("failed to prepare framebuffer\n");
+		return ret;
 	}
 
 	return 0;
@@ -689,8 +682,7 @@ static void dpu_plane_cleanup_fb(struct drm_plane *plane,
 
 	DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", old_state->fb->base.id);
 
-	msm_framebuffer_cleanup(old_state->fb, old_pstate->vm,
-				old_pstate->needs_dirtyfb);
+	msm_framebuffer_cleanup(old_state->fb, old_pstate->needs_dirtyfb);
 }
 
 static int dpu_plane_check_inline_rotation(struct dpu_plane *pdpu,
@@ -1457,7 +1449,7 @@ static void dpu_plane_sspp_atomic_update(struct drm_plane *plane,
 	pstate->needs_qos_remap |= (is_rt_pipe != pdpu->is_rt_pipe);
 	pdpu->is_rt_pipe = is_rt_pipe;
 
-	dpu_format_populate_addrs(pstate->vm, new_state->fb, &pstate->layout);
+	dpu_format_populate_addrs(new_state->fb, &pstate->layout);
 
 	DPU_DEBUG_PLANE(pdpu, "FB[%u] " DRM_RECT_FP_FMT "->crtc%u " DRM_RECT_FMT
 			", %p4cc ubwc %d\n", fb->base.id, DRM_RECT_FP_ARG(&state->src),
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
index 3578f52048a5..a3a6e9028333 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h
@@ -17,7 +17,6 @@
 /**
  * struct dpu_plane_state: Define dpu extension of drm plane state object
  * @base:	base drm plane state object
- * @vm:	pointer to address space for input/output buffers
  * @pipe:	software pipe description
  * @r_pipe:	software pipe description of the second pipe
  * @pipe_cfg:	software pipe configuration
@@ -34,7 +33,6 @@
  */
 struct dpu_plane_state {
 	struct drm_plane_state base;
-	struct msm_gem_vm *vm;
 	struct dpu_sw_pipe pipe;
 	struct dpu_sw_pipe r_pipe;
 	struct dpu_sw_pipe_cfg pipe_cfg;
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c
index 7743be6167f8..098c3b5ff2b2 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c
@@ -79,30 +79,25 @@ static const struct drm_plane_funcs mdp4_plane_funcs = {
 static int mdp4_plane_prepare_fb(struct drm_plane *plane,
 				 struct drm_plane_state *new_state)
 {
-	struct msm_drm_private *priv = plane->dev->dev_private;
-	struct msm_kms *kms = priv->kms;
-
 	if (!new_state->fb)
 		return 0;
 
 	drm_gem_plane_helper_prepare_fb(plane, new_state);
 
-	return msm_framebuffer_prepare(new_state->fb, kms->vm, false);
+	return msm_framebuffer_prepare(new_state->fb, false);
 }
 
 static void mdp4_plane_cleanup_fb(struct drm_plane *plane,
 				  struct drm_plane_state *old_state)
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
-	struct mdp4_kms *mdp4_kms = get_kms(plane);
-	struct msm_kms *kms = &mdp4_kms->base.base;
 	struct drm_framebuffer *fb = old_state->fb;
 
 	if (!fb)
 		return;
 
 	DBG("%s: cleanup: FB[%u]", mdp4_plane->name, fb->base.id);
-	msm_framebuffer_cleanup(fb, kms->vm, false);
+	msm_framebuffer_cleanup(fb, false);
 }
 
 
@@ -141,7 +136,6 @@ static void mdp4_plane_set_scanout(struct drm_plane *plane,
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
 	struct mdp4_kms *mdp4_kms = get_kms(plane);
-	struct msm_kms *kms = &mdp4_kms->base.base;
 	enum mdp4_pipe pipe = mdp4_plane->pipe;
 
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_STRIDE_A(pipe),
@@ -153,13 +147,13 @@ static void mdp4_plane_set_scanout(struct drm_plane *plane,
 			MDP4_PIPE_SRC_STRIDE_B_P3(fb->pitches[3]));
 
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP0_BASE(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 0));
+			msm_framebuffer_iova(fb, 0));
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP1_BASE(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 1));
+			msm_framebuffer_iova(fb, 1));
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP2_BASE(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 2));
+			msm_framebuffer_iova(fb, 2));
 	mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP3_BASE(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 3));
+			msm_framebuffer_iova(fb, 3));
 }
 
 static void mdp4_write_csc_config(struct mdp4_kms *mdp4_kms,
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c
index 9f68a4747203..7c790406d533 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c
@@ -135,8 +135,6 @@ static const struct drm_plane_funcs mdp5_plane_funcs = {
 static int mdp5_plane_prepare_fb(struct drm_plane *plane,
 				 struct drm_plane_state *new_state)
 {
-	struct msm_drm_private *priv = plane->dev->dev_private;
-	struct msm_kms *kms = priv->kms;
 	bool needs_dirtyfb = to_mdp5_plane_state(new_state)->needs_dirtyfb;
 
 	if (!new_state->fb)
@@ -144,14 +142,12 @@ static int mdp5_plane_prepare_fb(struct drm_plane *plane,
 
 	drm_gem_plane_helper_prepare_fb(plane, new_state);
 
-	return msm_framebuffer_prepare(new_state->fb, kms->vm, needs_dirtyfb);
+	return msm_framebuffer_prepare(new_state->fb, needs_dirtyfb);
 }
 
 static void mdp5_plane_cleanup_fb(struct drm_plane *plane,
 				  struct drm_plane_state *old_state)
 {
-	struct mdp5_kms *mdp5_kms = get_kms(plane);
-	struct msm_kms *kms = &mdp5_kms->base.base;
 	struct drm_framebuffer *fb = old_state->fb;
 	bool needed_dirtyfb = to_mdp5_plane_state(old_state)->needs_dirtyfb;
 
@@ -159,7 +155,7 @@ static void mdp5_plane_cleanup_fb(struct drm_plane *plane,
 		return;
 
 	DBG("%s: cleanup: FB[%u]", plane->name, fb->base.id);
-	msm_framebuffer_cleanup(fb, kms->vm, needed_dirtyfb);
+	msm_framebuffer_cleanup(fb, needed_dirtyfb);
 }
 
 static int mdp5_plane_atomic_check_with_state(struct drm_crtc_state *crtc_state,
@@ -467,8 +463,6 @@ static void set_scanout_locked(struct mdp5_kms *mdp5_kms,
 			       enum mdp5_pipe pipe,
 			       struct drm_framebuffer *fb)
 {
-	struct msm_kms *kms = &mdp5_kms->base.base;
-
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_STRIDE_A(pipe),
 			MDP5_PIPE_SRC_STRIDE_A_P0(fb->pitches[0]) |
 			MDP5_PIPE_SRC_STRIDE_A_P1(fb->pitches[1]));
@@ -478,13 +472,13 @@ static void set_scanout_locked(struct mdp5_kms *mdp5_kms,
 			MDP5_PIPE_SRC_STRIDE_B_P3(fb->pitches[3]));
 
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC0_ADDR(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 0));
+			msm_framebuffer_iova(fb, 0));
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC1_ADDR(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 1));
+			msm_framebuffer_iova(fb, 1));
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC2_ADDR(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 2));
+			msm_framebuffer_iova(fb, 2));
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC3_ADDR(pipe),
-			msm_framebuffer_iova(fb, kms->vm, 3));
+			msm_framebuffer_iova(fb, 3));
 }
 
 /* Note: mdp5_plane->pipe_lock must be locked */
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 761e7e221ad9..eb009bd193e3 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -274,12 +274,9 @@ struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev,
 int msm_gem_prime_pin(struct drm_gem_object *obj);
 void msm_gem_prime_unpin(struct drm_gem_object *obj);
 
-int msm_framebuffer_prepare(struct drm_framebuffer *fb,
-		struct msm_gem_vm *vm, bool needs_dirtyfb);
-void msm_framebuffer_cleanup(struct drm_framebuffer *fb,
-		struct msm_gem_vm *vm, bool needed_dirtyfb);
-uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb,
-		struct msm_gem_vm *vm, int plane);
+int msm_framebuffer_prepare(struct drm_framebuffer *fb, bool needs_dirtyfb);
+void msm_framebuffer_cleanup(struct drm_framebuffer *fb, bool needed_dirtyfb);
+uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, int plane);
 struct drm_gem_object *msm_framebuffer_bo(struct drm_framebuffer *fb, int plane);
 const struct msm_format *msm_framebuffer_format(struct drm_framebuffer *fb);
 struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev,
diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c
index 6df318b73534..8a3b88130f4d 100644
--- a/drivers/gpu/drm/msm/msm_fb.c
+++ b/drivers/gpu/drm/msm/msm_fb.c
@@ -75,10 +75,10 @@ void msm_framebuffer_describe(struct drm_framebuffer *fb, struct seq_file *m)
 
 /* prepare/pin all the fb's bo's for scanout.
  */
-int msm_framebuffer_prepare(struct drm_framebuffer *fb,
-		struct msm_gem_vm *vm,
-		bool needs_dirtyfb)
+int msm_framebuffer_prepare(struct drm_framebuffer *fb, bool needs_dirtyfb)
 {
+	struct msm_drm_private *priv = fb->dev->dev_private;
+	struct msm_gem_vm *vm = priv->kms->vm;
 	struct msm_framebuffer *msm_fb = to_msm_framebuffer(fb);
 	int ret, i, n = fb->format->num_planes;
 
@@ -98,10 +98,10 @@ int msm_framebuffer_prepare(struct drm_framebuffer *fb,
 	return 0;
 }
 
-void msm_framebuffer_cleanup(struct drm_framebuffer *fb,
-		struct msm_gem_vm *vm,
-		bool needed_dirtyfb)
+void msm_framebuffer_cleanup(struct drm_framebuffer *fb, bool needed_dirtyfb)
 {
+	struct msm_drm_private *priv = fb->dev->dev_private;
+	struct msm_gem_vm *vm = priv->kms->vm;
 	struct msm_framebuffer *msm_fb = to_msm_framebuffer(fb);
 	int i, n = fb->format->num_planes;
 
@@ -115,8 +115,7 @@ void msm_framebuffer_cleanup(struct drm_framebuffer *fb,
 		memset(msm_fb->iova, 0, sizeof(msm_fb->iova));
 }
 
-uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb,
-		struct msm_gem_vm *vm, int plane)
+uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, int plane)
 {
 	struct msm_framebuffer *msm_fb = to_msm_framebuffer(fb);
 	return msm_fb->iova[plane] + fb->offsets[plane];
-- 
2.50.0