[PATCH AUTOSEL 6.5 10/36] drm/amdgpu: Increase soft IH ring size

[PATCH AUTOSEL 6.5 10/36] drm/amdgpu: Increase soft IH ring size

[Sasha Levin]

From: Philip Yang <Philip.Yang@amd.com>

[ Upstream commit bf80d34b6c58ad1c4f76067ecd460a148eab9d39 ]

Retry faults are delegated to soft IH ring and then processed by
deferred worker. Current soft IH ring size PAGE_SIZE can store 128
entries, which may overflow and drop retry faults, causes HW stucks
because the retry fault is not recovered.

Increase soft IH ring size to 8KB, enough to store 256 CAM entries
because we clear the CAM entry after handling the retry fault from soft
ring.

Define macro IH_RING_SIZE and IH_SW_RING_SIZE to remove duplicate
constant.

Show warning message if soft IH ring overflows with CAM enabled because
this should not happen.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c  | 8 ++++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h  | 7 +++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/ih_v6_0.c    | 4 ++--
 drivers/gpu/drm/amd/amdgpu/navi10_ih.c  | 4 ++--
 drivers/gpu/drm/amd/amdgpu/vega10_ih.c  | 4 ++--
 drivers/gpu/drm/amd/amdgpu/vega20_ih.c  | 4 ++--
 7 files changed, 20 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
index fceb3b384955a..f3b0aaf3ebc69 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c
@@ -138,6 +138,7 @@ void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih)
 /**
  * amdgpu_ih_ring_write - write IV to the ring buffer
  *
+ * @adev: amdgpu_device pointer
  * @ih: ih ring to write to
  * @iv: the iv to write
  * @num_dw: size of the iv in dw
@@ -145,8 +146,8 @@ void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih)
  * Writes an IV to the ring buffer using the CPU and increment the wptr.
  * Used for testing and delegating IVs to a software ring.
  */
-void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv,
-			  unsigned int num_dw)
+void amdgpu_ih_ring_write(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
+			  const uint32_t *iv, unsigned int num_dw)
 {
 	uint32_t wptr = le32_to_cpu(*ih->wptr_cpu) >> 2;
 	unsigned int i;
@@ -161,6 +162,9 @@ void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv,
 	if (wptr != READ_ONCE(ih->rptr)) {
 		wmb();
 		WRITE_ONCE(*ih->wptr_cpu, cpu_to_le32(wptr));
+	} else if (adev->irq.retry_cam_enabled) {
+		dev_warn_once(adev->dev, "IH soft ring buffer overflow 0x%X, 0x%X\n",
+			      wptr, ih->rptr);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
index dd1c2eded6b9d..6c6184f0dbc17 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
@@ -27,6 +27,9 @@
 /* Maximum number of IVs processed at once */
 #define AMDGPU_IH_MAX_NUM_IVS	32
 
+#define IH_RING_SIZE	(256 * 1024)
+#define IH_SW_RING_SIZE	(8 * 1024)	/* enough for 256 CAM entries */
+
 struct amdgpu_device;
 struct amdgpu_iv_entry;
 
@@ -97,8 +100,8 @@ struct amdgpu_ih_funcs {
 int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
 			unsigned ring_size, bool use_bus_addr);
 void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih);
-void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv,
-			  unsigned int num_dw);
+void amdgpu_ih_ring_write(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih,
+			  const uint32_t *iv, unsigned int num_dw);
 int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev,
 					    struct amdgpu_ih_ring *ih);
 int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 5273decc5753b..fa6d0adcec206 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -493,7 +493,7 @@ void amdgpu_irq_delegate(struct amdgpu_device *adev,
 			 struct amdgpu_iv_entry *entry,
 			 unsigned int num_dw)
 {
-	amdgpu_ih_ring_write(&adev->irq.ih_soft, entry->iv_entry, num_dw);
+	amdgpu_ih_ring_write(adev, &adev->irq.ih_soft, entry->iv_entry, num_dw);
 	schedule_work(&adev->irq.ih_soft_work);
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
index b02e1cef78a76..980b241200803 100644
--- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
@@ -535,7 +535,7 @@ static int ih_v6_0_sw_init(void *handle)
 	 * use bus address for ih ring by psp bl */
 	use_bus_addr =
 		(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) ? false : true;
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr);
 	if (r)
 		return r;
 
@@ -548,7 +548,7 @@ static int ih_v6_0_sw_init(void *handle)
 	/* initialize ih control register offset */
 	ih_v6_0_init_register_offset(adev);
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, true);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
index eec13cb5bf758..b6a8478dabf43 100644
--- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c
@@ -565,7 +565,7 @@ static int navi10_ih_sw_init(void *handle)
 		use_bus_addr = false;
 	else
 		use_bus_addr = true;
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr);
 	if (r)
 		return r;
 
@@ -578,7 +578,7 @@ static int navi10_ih_sw_init(void *handle)
 	/* initialize ih control registers offset */
 	navi10_ih_init_register_offset(adev);
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, true);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
index 1e83db0c5438d..d364c6dd152c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
@@ -485,7 +485,7 @@ static int vega10_ih_sw_init(void *handle)
 	if (r)
 		return r;
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, true);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, true);
 	if (r)
 		return r;
 
@@ -510,7 +510,7 @@ static int vega10_ih_sw_init(void *handle)
 	/* initialize ih control registers offset */
 	vega10_ih_init_register_offset(adev);
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, true);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
index 4d719df376a72..544ee55a22da6 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c
@@ -539,7 +539,7 @@ static int vega20_ih_sw_init(void *handle)
 	    (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 2)))
 		use_bus_addr = false;
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr);
 	if (r)
 		return r;
 
@@ -565,7 +565,7 @@ static int vega20_ih_sw_init(void *handle)
 	/* initialize ih control registers offset */
 	vega20_ih_init_register_offset(adev);
 
-	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, use_bus_addr);
+	r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, use_bus_addr);
 	if (r)
 		return r;
 
-- 
2.40.1

[PATCH AUTOSEL 6.5 11/36] drm/amd/display: Add stream overhead in BW calculations for 128b/132b

[Sasha Levin]

From: George Shen <george.shen@amd.com>

[ Upstream commit 974764180838516f80a13257da67a1ec6afb87d4 ]

[Why]
Current BW calculations do not account for the additional padding added
for uncompressed pixel-to-symbol packing.

This results in X.Y being too low for 128b/132b SST streams in certain
scenarios. If X.Y is too low, end user can observe image corruption.

[How]
Add function to calculate stream overhead to timing BW calculation for
128b/132b SST cases.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Acked-by: Alan Liu <haoping.liu@amd.com>
Signed-off-by: George Shen <george.shen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/display/dc/dc.h         |  2 +
 drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c | 42 +++++++++++++++++++++
 2 files changed, 44 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 63948170fd6d9..3d4180f7ba349 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1409,6 +1409,8 @@ struct dc_plane_state *dc_get_surface_for_mpcc(struct dc *dc,
 
 uint32_t dc_get_opp_for_plane(struct dc *dc, struct dc_plane_state *plane);
 
+void dc_set_disable_128b_132b_stream_overhead(bool disable);
+
 /* The function returns minimum bandwidth required to drive a given timing
  * return - minimum required timing bandwidth in kbps.
  */
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
index 58dd62cce4bb9..aed0d3dafa247 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
@@ -40,6 +40,8 @@ static bool dsc_policy_enable_dsc_when_not_needed;
 
 static bool dsc_policy_disable_dsc_stream_overhead;
 
+static bool disable_128b_132b_stream_overhead;
+
 #ifndef MAX
 #define MAX(X, Y) ((X) > (Y) ? (X) : (Y))
 #endif
@@ -47,6 +49,41 @@ static bool dsc_policy_disable_dsc_stream_overhead;
 #define MIN(X, Y) ((X) < (Y) ? (X) : (Y))
 #endif
 
+/* Need to account for padding due to pixel-to-symbol packing
+ * for uncompressed 128b/132b streams.
+ */
+static uint32_t apply_128b_132b_stream_overhead(
+	const struct dc_crtc_timing *timing, const uint32_t kbps)
+{
+	uint32_t total_kbps = kbps;
+
+	if (disable_128b_132b_stream_overhead)
+		return kbps;
+
+	if (!timing->flags.DSC) {
+		struct fixed31_32 bpp;
+		struct fixed31_32 overhead_factor;
+
+		bpp = dc_fixpt_from_int(kbps);
+		bpp = dc_fixpt_div_int(bpp, timing->pix_clk_100hz / 10);
+
+		/* Symbols_per_HActive = HActive * bpp / (4 lanes * 32-bit symbol size)
+		 * Overhead_factor = ceil(Symbols_per_HActive) / Symbols_per_HActive
+		 */
+		overhead_factor = dc_fixpt_from_int(timing->h_addressable);
+		overhead_factor = dc_fixpt_mul(overhead_factor, bpp);
+		overhead_factor = dc_fixpt_div_int(overhead_factor, 128);
+		overhead_factor = dc_fixpt_div(
+			dc_fixpt_from_int(dc_fixpt_ceil(overhead_factor)),
+			overhead_factor);
+
+		total_kbps = dc_fixpt_ceil(
+			dc_fixpt_mul_int(overhead_factor, total_kbps));
+	}
+
+	return total_kbps;
+}
+
 uint32_t dc_bandwidth_in_kbps_from_timing(
 	const struct dc_crtc_timing *timing)
 {
@@ -1165,6 +1202,11 @@ void dc_dsc_policy_set_disable_dsc_stream_overhead(bool disable)
 	dsc_policy_disable_dsc_stream_overhead = disable;
 }
 
+void dc_set_disable_128b_132b_stream_overhead(bool disable)
+{
+	disable_128b_132b_stream_overhead = disable;
+}
+
 void dc_dsc_get_default_config_option(const struct dc *dc, struct dc_dsc_config_options *options)
 {
 	options->dsc_min_slice_height_override = dc->debug.dsc_min_slice_height_override;
-- 
2.40.1

[PATCH AUTOSEL 6.5 13/36] drm/amdgpu: Update ring scheduler info as needed

[Sasha Levin]

From: Lijo Lazar <lijo.lazar@amd.com>

[ Upstream commit 6cb209ed68e45c8e4b71d97a037ac6b7dbce9b50 ]

Not all rings have scheduler associated. Only update scheduler data for
rings with scheduler. It could result in out of bound access as total
rings are more than those associated with particular IPs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
index 72b629a78c62c..d0fc62784e821 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c
@@ -134,7 +134,7 @@ static int aqua_vanjaram_xcp_sched_list_update(
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
 		ring = adev->rings[i];
-		if (!ring || !ring->sched.ready)
+		if (!ring || !ring->sched.ready || ring->no_scheduler)
 			continue;
 
 		aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id);
-- 
2.40.1

[PATCH AUTOSEL 6.5 14/36] drm/amd/display: Read down-spread percentage from lut to adjust dprefclk.

[Sasha Levin]

From: Martin Tsai <martin.tsai@amd.com>

[ Upstream commit 6917b0b711713b9d84d7e0844e9aa613997a51b2 ]

[Why]
Panels show corruption with high refresh rate timings when
ss is enabled.

[How]
Read down-spread percentage from lut to adjust dprefclk.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Martin Tsai <martin.tsai@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.c        | 33 +++++++++++++++++--
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.h        |  7 ++++
 2 files changed, 38 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 2f7c8996b19d3..3ba2e13d691d9 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -87,6 +87,14 @@ static const struct IP_BASE CLK_BASE = { { { { 0x00016C00, 0x02401800, 0, 0, 0,
 #define CLK1_CLK_PLL_REQ__PllSpineDiv_MASK	0x0000F000L
 #define CLK1_CLK_PLL_REQ__FbMult_frac_MASK	0xFFFF0000L
 
+#define regCLK1_CLK2_BYPASS_CNTL			0x029c
+#define regCLK1_CLK2_BYPASS_CNTL_BASE_IDX	0
+
+#define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_SEL__SHIFT	0x0
+#define CLK1_CLK2_BYPASS_CNTL__LK2_BYPASS_DIV__SHIFT	0x10
+#define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_SEL_MASK		0x00000007L
+#define CLK1_CLK2_BYPASS_CNTL__LK2_BYPASS_DIV_MASK		0x000F0000L
+
 #define REG(reg_name) \
 	(CLK_BASE.instance[0].segment[reg ## reg_name ## _BASE_IDX] + reg ## reg_name)
 
@@ -436,6 +444,11 @@ static DpmClocks314_t dummy_clocks;
 
 static struct dcn314_watermarks dummy_wms = { 0 };
 
+static struct dcn314_ss_info_table ss_info_table = {
+	.ss_divider = 1000,
+	.ss_percentage = {0, 0, 375, 375, 375}
+};
+
 static void dcn314_build_watermark_ranges(struct clk_bw_params *bw_params, struct dcn314_watermarks *table)
 {
 	int i, num_valid_sets;
@@ -715,6 +728,20 @@ static struct clk_mgr_funcs dcn314_funcs = {
 };
 extern struct clk_mgr_funcs dcn3_fpga_funcs;
 
+static void dcn314_read_ss_info_from_lut(struct clk_mgr_internal *clk_mgr)
+{
+	uint32_t clock_source;
+
+	REG_GET(CLK1_CLK2_BYPASS_CNTL, CLK2_BYPASS_SEL, &clock_source);
+
+	clk_mgr->dprefclk_ss_percentage = ss_info_table.ss_percentage[clock_source];
+
+	if (clk_mgr->dprefclk_ss_percentage != 0) {
+		clk_mgr->ss_on_dprefclk = true;
+		clk_mgr->dprefclk_ss_divider = ss_info_table.ss_divider;
+	}
+}
+
 void dcn314_clk_mgr_construct(
 		struct dc_context *ctx,
 		struct clk_mgr_dcn314 *clk_mgr,
@@ -781,9 +808,11 @@ void dcn314_clk_mgr_construct(
 
 	clk_mgr->base.base.dprefclk_khz = 600000;
 	clk_mgr->base.base.clks.ref_dtbclk_khz = 600000;
-	dce_clock_read_ss_info(&clk_mgr->base);
+
+	dcn314_read_ss_info_from_lut(&clk_mgr->base);
 	/*if bios enabled SS, driver needs to adjust dtb clock, only enable with correct bios*/
-	//clk_mgr->base.dccg->ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(clk_mgr_internal, clk_mgr->base.base.dprefclk_khz);
+	clk_mgr->base.base.dprefclk_khz =
+		dce_adjust_dp_ref_freq_for_ss(&clk_mgr->base, clk_mgr->base.base.dprefclk_khz);
 
 	clk_mgr->base.base.bw_params = &dcn314_bw_params;
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
index 171f84340eb2f..e0670dafe2600 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
@@ -28,6 +28,8 @@
 #define __DCN314_CLK_MGR_H__
 #include "clk_mgr_internal.h"
 
+#define NUM_CLOCK_SOURCES   5
+
 struct dcn314_watermarks;
 
 struct dcn314_smu_watermark_set {
@@ -40,6 +42,11 @@ struct clk_mgr_dcn314 {
 	struct dcn314_smu_watermark_set smu_wm_set;
 };
 
+struct dcn314_ss_info_table {
+	uint32_t ss_divider;
+	uint32_t ss_percentage[NUM_CLOCK_SOURCES];
+};
+
 bool dcn314_are_clock_states_equal(struct dc_clocks *a,
 		struct dc_clocks *b);
 
-- 
2.40.1

[PATCH AUTOSEL 6.5 15/36] drm/amd/display: Fix underflow issue on 175hz timing

[Sasha Levin]

From: Leo Ma <hanghong.ma@amd.com>

[ Upstream commit 735688eb905db529efea0c78466fccc1461c3fde ]

[Why]
Screen underflows happen on 175hz timing for 3 plane overlay case.

[How]
Based on dst y prefetch value clamp to equ or oto for bandwidth
calculation.

Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Leo Ma <hanghong.ma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../amd/display/dc/dml/dcn32/display_mode_vba_util_32.c    | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
index a50e7f4dce421..f74e5fc8218f5 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c
@@ -3459,6 +3459,7 @@ bool dml32_CalculatePrefetchSchedule(
 	double TimeForFetchingMetaPTE = 0;
 	double TimeForFetchingRowInVBlank = 0;
 	double LinesToRequestPrefetchPixelData = 0;
+	double LinesForPrefetchBandwidth = 0;
 	unsigned int HostVMDynamicLevelsTrips;
 	double  trip_to_mem;
 	double  Tvm_trips;
@@ -3888,11 +3889,15 @@ bool dml32_CalculatePrefetchSchedule(
 			TimeForFetchingMetaPTE = Tvm_oto;
 			TimeForFetchingRowInVBlank = Tr0_oto;
 			*PrefetchBandwidth = prefetch_bw_oto;
+			/* Clamp to oto for bandwidth calculation */
+			LinesForPrefetchBandwidth = dst_y_prefetch_oto;
 		} else {
 			*DestinationLinesForPrefetch = dst_y_prefetch_equ;
 			TimeForFetchingMetaPTE = Tvm_equ;
 			TimeForFetchingRowInVBlank = Tr0_equ;
 			*PrefetchBandwidth = prefetch_bw_equ;
+			/* Clamp to equ for bandwidth calculation */
+			LinesForPrefetchBandwidth = dst_y_prefetch_equ;
 		}
 
 		*DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * TimeForFetchingMetaPTE / LineTime, 1.0) / 4.0;
@@ -3900,7 +3905,7 @@ bool dml32_CalculatePrefetchSchedule(
 		*DestinationLinesToRequestRowInVBlank =
 				dml_ceil(4.0 * TimeForFetchingRowInVBlank / LineTime, 1.0) / 4.0;
 
-		LinesToRequestPrefetchPixelData = *DestinationLinesForPrefetch -
+		LinesToRequestPrefetchPixelData = LinesForPrefetchBandwidth -
 				*DestinationLinesToRequestVMInVBlank - 2 * *DestinationLinesToRequestRowInVBlank;
 
 #ifdef __DML_VBA_DEBUG__
-- 
2.40.1

[PATCH AUTOSEL 6.5 20/36] drm/amd/display: Use DTBCLK as refclk instead of DPREFCLK

[Sasha Levin]

From: Austin Zheng <austin.zheng@amd.com>

[ Upstream commit 4a30cc2bd281fa176a68b5305cd3695d636152ad ]

[Why]
Flash of corruption observed when UCLK switching after transitioning
from DTBCLK to DPREFCLK on subVP(DP) + subVP(HDMI) config
Scenario where DPREFCLK is required instead of DTBCLK is not expected

[How]
Always set the DTBCLK source as DTBCLK0

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Austin Zheng <austin.zheng@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
index 61ceff6bc0b19..921f58c0c729b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c
@@ -281,7 +281,8 @@ static void dccg32_set_dpstreamclk(
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
 	/* set the dtbclk_p source */
-	dccg32_set_dtbclk_p_src(dccg, src, otg_inst);
+	/* always program refclk as DTBCLK. No use-case expected to require DPREFCLK as refclk */
+	dccg32_set_dtbclk_p_src(dccg, DTBCLK0, otg_inst);
 
 	/* enabled to select one of the DTBCLKs for pipe */
 	switch (dp_hpo_inst) {
-- 
2.40.1

[PATCH AUTOSEL 6.5 21/36] drm/amd/display: Blocking invalid 420 modes on HDMI TMDS for DCN31

[Sasha Levin]

From: Leo Chen <sancchen@amd.com>

[ Upstream commit 026a71babf48efb6b9884a3a66fa31aec9e1ea54 ]

[Why & How]
HDMI TMDS does not have ODM support. Filtering 420 modes that
exceed the 4096 FMT limitation on DCN31 will resolve
intermittent corruptions issues.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c    | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
index 43016c462251f..9d996d5fc3ffa 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c
@@ -4135,7 +4135,9 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 				}
 				if (v->OutputFormat[k] == dm_420 && v->HActive[k] > DCN31_MAX_FMT_420_BUFFER_WIDTH
 						&& v->ODMCombineEnablePerState[i][k] != dm_odm_combine_mode_4to1) {
-					if (v->HActive[k] / 2 > DCN31_MAX_FMT_420_BUFFER_WIDTH) {
+					if (v->Output[k] == dm_hdmi) {
+						FMTBufferExceeded = true;
+					} else if (v->HActive[k] / 2 > DCN31_MAX_FMT_420_BUFFER_WIDTH) {
 						v->ODMCombineEnablePerState[i][k] = dm_odm_combine_mode_4to1;
 						v->PlaneRequiredDISPCLK = v->PlaneRequiredDISPCLKWithODMCombine4To1;
 
-- 
2.40.1

[PATCH AUTOSEL 6.5 22/36] drm/amd/display: Blocking invalid 420 modes on HDMI TMDS for DCN314

[Sasha Levin]

From: Leo Chen <sancchen@amd.com>

[ Upstream commit 4c6107a653ccf361cb1b6ba35d558a1a5e6e57ac ]

[Why & How]
HDMI TMDS does not have ODM support. Filtering 420 modes that
exceed the 4096 FMT limitation on DCN314 will resolve
intermittent corruptions issues.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 .../gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c  | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
index 9010c47476e92..b763786bfcc85 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c
@@ -4227,7 +4227,9 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_
 				}
 				if (v->OutputFormat[k] == dm_420 && v->HActive[k] > DCN314_MAX_FMT_420_BUFFER_WIDTH
 						&& v->ODMCombineEnablePerState[i][k] != dm_odm_combine_mode_4to1) {
-					if (v->HActive[k] / 2 > DCN314_MAX_FMT_420_BUFFER_WIDTH) {
+					if (v->Output[k] == dm_hdmi) {
+						FMTBufferExceeded = true;
+					} else if (v->HActive[k] / 2 > DCN314_MAX_FMT_420_BUFFER_WIDTH) {
 						v->ODMCombineEnablePerState[i][k] = dm_odm_combine_mode_4to1;
 						v->PlaneRequiredDISPCLK = v->PlaneRequiredDISPCLKWithODMCombine4To1;
 
-- 
2.40.1

[PATCH AUTOSEL 6.5 23/36] drm/amd/display: Use max memclk variable when setting max memclk

[Sasha Levin]

From: Alvin Lee <alvin.lee2@amd.com>

[ Upstream commit 2b1b838ea8e5437ef06a29818d16e9efdfaf0037 ]

[Description]
In overclocking scenarios the max memclk could be higher
than the DC mode limit. However, for configs that don't
support MCLK switching we need to set the max memclk to
the overclocked max instead of the DC mode max or we
could result in underflow.

Reviewed-by: Samson Tam <samson.tam@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index cb992aca760dc..5fc78bf927bbc 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -802,7 +802,7 @@ static void dcn32_set_hard_min_memclk(struct clk_mgr *clk_mgr_base, bool current
 					khz_to_mhz_ceil(clk_mgr_base->clks.dramclk_khz));
 		else
 			dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK,
-					clk_mgr_base->bw_params->clk_table.entries[clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_memclk_levels - 1].memclk_mhz);
+					clk_mgr_base->bw_params->max_memclk_mhz);
 	} else {
 		dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK,
 				clk_mgr_base->bw_params->clk_table.entries[0].memclk_mhz);
-- 
2.40.1