[PATCH 00/25] Enable dislay support for Battlemage

[PATCH 00/25] Enable dislay support for Battlemage

[Balasubramani Vivekanandan]

Adds display support for Battlemage.
Reuses the patch "drm/xe/bmg: Add BMG platform definition" from the
patch series <link> to help build this series. So that review on this
series can continue without blocking on <link>.

Ankit Nautiyal (1):
  Revert "drm/i915/dgfx: DGFX uses direct VBT pin mapping"

Anusha Srivatsa (1):
  drm/i915/xe2hpd: Add missing chicken bit register programming

Balasubramani Vivekanandan (9):
  drm/i915/display: Prepare to handle new C20 PLL register address
  drm/xe/bmg: Define IS_BATTLEMAGE macro
  drm/i915/bmg: Define IS_BATTLEMAGE macro
  drm/i915/xe2: Skip CCS modifiers for Xe2 platforms
  drm/i915/xe2hpd: Add new C20 PLL register address
  drm/i915/xe2hpd: Add support for eDP PLL configuration
  drm/xe/xe2hpd: Define a new DRAM type INTEL_DRAM_GDDR
  drm/i915/xe2hpd: Set maximum DP rate to UHBR13.5
  drm/xe/bmg: Enable the display support

Clint Taylor (2):
  drm/i915/xe2hpd: Initial cdclk table
  drm/xe/display: Lane reversal requires writes to both context lanes

José Roberto de Souza (2):
  drm/i915/xe2hpd: Properly disable power in port A
  drm/i915/xe2hpd: Do not program MBUS_DBOX BW credits

Lucas De Marchi (1):
  drm/i915/xe2hpd: Add display info

Matt Roper (3):
  drm/xe/bmg: Add BMG platform definition
  drm/i915/xe2hpd: Add max memory bandwidth algorithm
  drm/i915/bmg: BMG should re-use MTL's south display logic

Matthew Auld (2):
  drm/xe/gt_print: add xe_gt_err_once()
  drm/i915/display: perform transient flush

Mitul Golani (1):
  drm/i915/display: Enable RM timeout detection

Nirmoy Das (1):
  drm/xe/device: implement transient flush

Radhakrishna Sripada (1):
  drm/i915/bmg: Extend DG2 tc check to future

Ravi Kumar Vodapalli (1):
  drm/i915/xe2hpd: update pll values in sync with Bspec

 drivers/gpu/drm/i915/display/intel_bios.c     |   5 +-
 drivers/gpu/drm/i915/display/intel_bw.c       |  65 +++-
 drivers/gpu/drm/i915/display/intel_cdclk.c    |  11 +
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 297 ++++++++++++++++--
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |  45 ++-
 drivers/gpu/drm/i915/display/intel_display.c  |  10 +-
 .../drm/i915/display/intel_display_device.c   |  16 +
 .../gpu/drm/i915/display/intel_display_irq.c  |  10 +
 .../drm/i915/display/intel_display_power.c    |   4 +
 drivers/gpu/drm/i915/display/intel_dp.c       |   3 +
 drivers/gpu/drm/i915/display/intel_fb.c       |  14 +-
 .../gpu/drm/i915/display/intel_frontbuffer.c  |   2 +
 drivers/gpu/drm/i915/display/intel_tdf.h      |  25 ++
 drivers/gpu/drm/i915/display/skl_watermark.c  |   2 +-
 drivers/gpu/drm/i915/i915_drv.h               |  11 +-
 drivers/gpu/drm/i915/i915_reg.h               |   4 +
 drivers/gpu/drm/i915/soc/intel_dram.c         |   4 +
 drivers/gpu/drm/i915/soc/intel_pch.c          |   4 +-
 drivers/gpu/drm/xe/Makefile                   |   3 +-
 .../gpu/drm/xe/compat-i915-headers/i915_drv.h |   1 +
 drivers/gpu/drm/xe/display/xe_tdf.c           |  13 +
 drivers/gpu/drm/xe/regs/xe_gt_regs.h          |   3 +
 drivers/gpu/drm/xe/xe_device.c                |  49 +++
 drivers/gpu/drm/xe/xe_device.h                |   2 +
 drivers/gpu/drm/xe/xe_device_types.h          |   1 +
 drivers/gpu/drm/xe/xe_gt_printk.h             |   3 +
 drivers/gpu/drm/xe/xe_pci.c                   |   8 +
 drivers/gpu/drm/xe/xe_platform_types.h        |   1 +
 include/drm/xe_pciids.h                       |   7 +
 29 files changed, 571 insertions(+), 52 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_tdf.h
 create mode 100644 drivers/gpu/drm/xe/display/xe_tdf.c

-- 
2.25.1

[PATCH 01/25] drm/i915/display: Prepare to handle new C20 PLL register address

[Balasubramani Vivekanandan]

New platforms have different addresses for C20 PLL registers. This patch
prepares the driver to work with different register addresses.
New structure `struct intel_c20pll_reg` is created to hold the register
addresses for each platform with different register address.

CC: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 53 +++++++++++++------
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h | 36 ++++++++++---
 2 files changed, 65 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index a2c4bf33155f..13a2e3db2812 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -759,6 +759,17 @@ static const struct intel_c10pll_state * const mtl_c10_edp_tables[] = {
 	NULL,
 };
 
+static struct intel_c20pll_reg mtl_c20_reg = {
+	.tx_cnt_a = MTL_C20_A_TX_CNTX_CFG_ADDR,
+	.tx_cnt_b = MTL_C20_B_TX_CNTX_CFG_ADDR,
+	.cmn_cnt_a = MTL_C20_A_CMN_CNTX_CFG_ADDR,
+	.cmn_cnt_b = MTL_C20_B_CMN_CNTX_CFG_ADDR,
+	.mplla_a = MTL_C20_A_MPLLA_CFG_ADDR,
+	.mplla_b = MTL_C20_B_MPLLA_CFG_ADDR,
+	.mpllb_a = MTL_C20_A_MPLLB_CFG_ADDR,
+	.mpllb_b = MTL_C20_B_MPLLB_CFG_ADDR
+};
+
 /* C20 basic DP 1.4 tables */
 static const struct intel_c20pll_state mtl_c20_dp_rbr = {
 	.clock = 162000,
@@ -2161,6 +2172,7 @@ static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 	bool cntx;
 	intel_wakeref_t wakeref;
 	int i;
+	struct intel_c20pll_reg *pll_reg = &mtl_c20_reg;
 
 	wakeref = intel_cx0_phy_transaction_begin(encoder);
 
@@ -2171,20 +2183,20 @@ static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 	for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) {
 		if (cntx)
 			pll_state->tx[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-							       PHY_C20_B_TX_CNTX_CFG(i));
+							       PHY_C20_B_TX_CNTX_CFG(pll_reg, i));
 		else
 			pll_state->tx[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-							       PHY_C20_A_TX_CNTX_CFG(i));
+							       PHY_C20_A_TX_CNTX_CFG(pll_reg, i));
 	}
 
 	/* Read common configuration */
 	for (i = 0; i < ARRAY_SIZE(pll_state->cmn); i++) {
 		if (cntx)
 			pll_state->cmn[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-								PHY_C20_B_CMN_CNTX_CFG(i));
+								PHY_C20_B_CMN_CNTX_CFG(pll_reg, i));
 		else
 			pll_state->cmn[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-								PHY_C20_A_CMN_CNTX_CFG(i));
+								PHY_C20_A_CMN_CNTX_CFG(pll_reg, i));
 	}
 
 	if (intel_c20phy_use_mpllb(pll_state)) {
@@ -2192,20 +2204,20 @@ static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 		for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) {
 			if (cntx)
 				pll_state->mpllb[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-									  PHY_C20_B_MPLLB_CNTX_CFG(i));
+									  PHY_C20_B_MPLLB_CNTX_CFG(pll_reg, i));
 			else
 				pll_state->mpllb[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-									  PHY_C20_A_MPLLB_CNTX_CFG(i));
+									  PHY_C20_A_MPLLB_CNTX_CFG(pll_reg, i));
 		}
 	} else {
 		/* MPLLA configuration */
 		for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) {
 			if (cntx)
 				pll_state->mplla[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-									  PHY_C20_B_MPLLA_CNTX_CFG(i));
+									  PHY_C20_B_MPLLA_CNTX_CFG(pll_reg, i));
 			else
 				pll_state->mplla[i] = intel_c20_sram_read(encoder, INTEL_CX0_LANE0,
-									  PHY_C20_A_MPLLA_CNTX_CFG(i));
+									  PHY_C20_A_MPLLA_CNTX_CFG(pll_reg, i));
 		}
 	}
 
@@ -2341,6 +2353,7 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 	u32 clock = crtc_state->port_clock;
 	bool cntx;
 	int i;
+	const struct intel_c20pll_reg *pll_reg = &mtl_c20_reg;
 
 	if (intel_crtc_has_dp_encoder(crtc_state))
 		dp = true;
@@ -2363,17 +2376,25 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 	/* 3.1 Tx configuration */
 	for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) {
 		if (cntx)
-			intel_c20_sram_write(encoder, INTEL_CX0_LANE0, PHY_C20_A_TX_CNTX_CFG(i), pll_state->tx[i]);
+			intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
+					     PHY_C20_A_TX_CNTX_CFG(pll_reg, i),
+					     pll_state->tx[i]);
 		else
-			intel_c20_sram_write(encoder, INTEL_CX0_LANE0, PHY_C20_B_TX_CNTX_CFG(i), pll_state->tx[i]);
+			intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
+					     PHY_C20_B_TX_CNTX_CFG(pll_reg, i),
+					     pll_state->tx[i]);
 	}
 
 	/* 3.2 common configuration */
 	for (i = 0; i < ARRAY_SIZE(pll_state->cmn); i++) {
 		if (cntx)
-			intel_c20_sram_write(encoder, INTEL_CX0_LANE0, PHY_C20_A_CMN_CNTX_CFG(i), pll_state->cmn[i]);
+			intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
+					     PHY_C20_A_CMN_CNTX_CFG(pll_reg, i),
+					     pll_state->cmn[i]);
 		else
-			intel_c20_sram_write(encoder, INTEL_CX0_LANE0, PHY_C20_B_CMN_CNTX_CFG(i), pll_state->cmn[i]);
+			intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
+					     PHY_C20_B_CMN_CNTX_CFG(pll_reg, i),
+					     pll_state->cmn[i]);
 	}
 
 	/* 3.3 mpllb or mplla configuration */
@@ -2381,22 +2402,22 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 		for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) {
 			if (cntx)
 				intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
-						     PHY_C20_A_MPLLB_CNTX_CFG(i),
+						     PHY_C20_A_MPLLB_CNTX_CFG(pll_reg, i),
 						     pll_state->mpllb[i]);
 			else
 				intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
-						     PHY_C20_B_MPLLB_CNTX_CFG(i),
+						     PHY_C20_B_MPLLB_CNTX_CFG(pll_reg, i),
 						     pll_state->mpllb[i]);
 		}
 	} else {
 		for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) {
 			if (cntx)
 				intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
-						     PHY_C20_A_MPLLA_CNTX_CFG(i),
+						     PHY_C20_A_MPLLA_CNTX_CFG(pll_reg, i),
 						     pll_state->mplla[i]);
 			else
 				intel_c20_sram_write(encoder, INTEL_CX0_LANE0,
-						     PHY_C20_B_MPLLA_CNTX_CFG(i),
+						     PHY_C20_B_MPLLA_CNTX_CFG(pll_reg, i),
 						     pll_state->mplla[i]);
 		}
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index bdd0c8c4ef97..882b98dc347b 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -254,24 +254,44 @@
 #define PHY_C20_VDR_CUSTOM_WIDTH	0xD02
 #define   PHY_C20_CUSTOM_WIDTH_MASK	REG_GENMASK(1, 0)
 #define   PHY_C20_CUSTOM_WIDTH(val)	REG_FIELD_PREP8(PHY_C20_CUSTOM_WIDTH_MASK, val)
-#define PHY_C20_A_TX_CNTX_CFG(idx)	(0xCF2E - (idx))
-#define PHY_C20_B_TX_CNTX_CFG(idx)	(0xCF2A - (idx))
+#define PHY_C20_A_TX_CNTX_CFG(reg, idx)		((reg)->tx_cnt_a - (idx))
+#define PHY_C20_B_TX_CNTX_CFG(reg, idx)		((reg)->tx_cnt_b - (idx))
 #define   C20_PHY_TX_RATE		REG_GENMASK(2, 0)
-#define PHY_C20_A_CMN_CNTX_CFG(idx)	(0xCDAA - (idx))
-#define PHY_C20_B_CMN_CNTX_CFG(idx)	(0xCDA5 - (idx))
-#define PHY_C20_A_MPLLA_CNTX_CFG(idx)	(0xCCF0 - (idx))
-#define PHY_C20_B_MPLLA_CNTX_CFG(idx)	(0xCCE5 - (idx))
+#define PHY_C20_A_CMN_CNTX_CFG(reg, idx)	((reg)->cmn_cnt_a - (idx))
+#define PHY_C20_B_CMN_CNTX_CFG(reg, idx)	((reg)->cmn_cnt_b - (idx))
+#define PHY_C20_A_MPLLA_CNTX_CFG(reg, idx)	((reg)->mplla_a - (idx))
+#define PHY_C20_B_MPLLA_CNTX_CFG(reg, idx)	((reg)->mplla_b - (idx))
 #define   C20_MPLLA_FRACEN		REG_BIT(14)
 #define   C20_FB_CLK_DIV4_EN		REG_BIT(13)
 #define   C20_MPLLA_TX_CLK_DIV_MASK	REG_GENMASK(10, 8)
-#define PHY_C20_A_MPLLB_CNTX_CFG(idx)	(0xCB5A - (idx))
-#define PHY_C20_B_MPLLB_CNTX_CFG(idx)	(0xCB4E - (idx))
+#define PHY_C20_A_MPLLB_CNTX_CFG(reg, idx)	((reg)->mpllb_a - (idx))
+#define PHY_C20_B_MPLLB_CNTX_CFG(reg, idx)	((reg)->mpllb_b - (idx))
 #define   C20_MPLLB_TX_CLK_DIV_MASK	REG_GENMASK(15, 13)
 #define   C20_MPLLB_FRACEN		REG_BIT(13)
 #define   C20_REF_CLK_MPLLB_DIV_MASK	REG_GENMASK(12, 10)
 #define   C20_MULTIPLIER_MASK		REG_GENMASK(11, 0)
 #define   C20_PHY_USE_MPLLB		REG_BIT(7)
 
+struct intel_c20pll_reg {
+	u16 tx_cnt_a;
+	u16 tx_cnt_b;
+	u16 cmn_cnt_a;
+	u16 cmn_cnt_b;
+	u16 mplla_a;
+	u16 mplla_b;
+	u16 mpllb_a;
+	u16 mpllb_b;
+};
+
+#define MTL_C20_A_TX_CNTX_CFG_ADDR	0xCF2E
+#define MTL_C20_B_TX_CNTX_CFG_ADDR	0xCF2A
+#define MTL_C20_A_CMN_CNTX_CFG_ADDR	0xCDAA
+#define MTL_C20_B_CMN_CNTX_CFG_ADDR	0xCDA5
+#define MTL_C20_A_MPLLA_CFG_ADDR	0xCCF0
+#define MTL_C20_B_MPLLA_CFG_ADDR	0xCCE5
+#define MTL_C20_A_MPLLB_CFG_ADDR	0xCB5A
+#define MTL_C20_B_MPLLB_CFG_ADDR	0xCB4E
+
 /* C20 Phy VSwing Masks */
 #define C20_PHY_VSWING_PREEMPH_MASK	REG_GENMASK8(5, 0)
 #define C20_PHY_VSWING_PREEMPH(val)	REG_FIELD_PREP8(C20_PHY_VSWING_PREEMPH_MASK, val)
-- 
2.25.1

[PATCH 02/25] drm/xe/bmg: Add BMG platform definition

[Balasubramani Vivekanandan]

From: Matt Roper <matthew.d.roper@intel.com>

BMG is a discrete GPU based on the Xe2 architecture.

Bspec: 68090
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/xe/xe_pci.c            | 7 +++++++
 drivers/gpu/drm/xe/xe_platform_types.h | 1 +
 include/drm/xe_pciids.h                | 7 +++++++
 3 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index c47ab4b67467..b3158053baee 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -337,6 +337,12 @@ static const struct xe_device_desc lnl_desc = {
 	.require_force_probe = true,
 };
 
+static const struct xe_device_desc bmg_desc = {
+	DGFX_FEATURES,
+	PLATFORM(XE_BATTLEMAGE),
+	.require_force_probe = true,
+};
+
 #undef PLATFORM
 __diag_pop();
 
@@ -379,6 +385,7 @@ static const struct pci_device_id pciidlist[] = {
 	XE_PVC_IDS(INTEL_VGA_DEVICE, &pvc_desc),
 	XE_MTL_IDS(INTEL_VGA_DEVICE, &mtl_desc),
 	XE_LNL_IDS(INTEL_VGA_DEVICE, &lnl_desc),
+	XE_BMG_IDS(INTEL_VGA_DEVICE, &bmg_desc),
 	{ }
 };
 MODULE_DEVICE_TABLE(pci, pciidlist);
diff --git a/drivers/gpu/drm/xe/xe_platform_types.h b/drivers/gpu/drm/xe/xe_platform_types.h
index 553f53dbd093..79b7042c4534 100644
--- a/drivers/gpu/drm/xe/xe_platform_types.h
+++ b/drivers/gpu/drm/xe/xe_platform_types.h
@@ -22,6 +22,7 @@ enum xe_platform {
 	XE_PVC,
 	XE_METEORLAKE,
 	XE_LUNARLAKE,
+	XE_BATTLEMAGE,
 };
 
 enum xe_subplatform {
diff --git a/include/drm/xe_pciids.h b/include/drm/xe_pciids.h
index c7fc288dacee..73d972a8aca1 100644
--- a/include/drm/xe_pciids.h
+++ b/include/drm/xe_pciids.h
@@ -208,4 +208,11 @@
 	MACRO__(0x64A0, ## __VA_ARGS__), \
 	MACRO__(0x64B0, ## __VA_ARGS__)
 
+#define XE_BMG_IDS(MACRO__, ...) \
+	MACRO__(0xE202, ## __VA_ARGS__), \
+	MACRO__(0xE20B, ## __VA_ARGS__), \
+	MACRO__(0xE20C, ## __VA_ARGS__), \
+	MACRO__(0xE20D, ## __VA_ARGS__), \
+	MACRO__(0xE212, ## __VA_ARGS__)
+
 #endif
-- 
2.25.1

[PATCH 03/25] drm/xe/bmg: Define IS_BATTLEMAGE macro

[Balasubramani Vivekanandan]

Common display code requires IS_BATTLEMAGE macro. Defined the macro.

Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
index a01d1b869c2d..9161d1fdf239 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
@@ -88,6 +88,7 @@ static inline struct drm_i915_private *kdev_to_i915(struct device *kdev)
 #define IS_DG2(dev_priv)	IS_PLATFORM(dev_priv, XE_DG2)
 #define IS_METEORLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_METEORLAKE)
 #define IS_LUNARLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_LUNARLAKE)
+#define IS_BATTLEMAGE(dev_priv)  IS_PLATFORM(dev_priv, XE_BATTLEMAGE)
 
 #define IS_HASWELL_ULT(dev_priv) (dev_priv && 0)
 #define IS_BROADWELL_ULT(dev_priv) (dev_priv && 0)
-- 
2.25.1

[PATCH 04/25] drm/i915/bmg: Define IS_BATTLEMAGE macro

[Balasubramani Vivekanandan]

Display code uses IS_BATTLEMAGE macro but the platform support doesn't
still exist in i915. So fake IS_BATTLEMAGE macro defined to enable
building i915 code.  We should make sure the macro parameter is used in
the always-false expression so that we don't run into "unused variable"
warnings from i915 builds if the IS_BATTLEMAGE() check is the only place
the i915 pointer gets used in a function.

While we're at it, also update the IS_LUNARLAKE macro to include the
parameter in the false expression for consistency.

Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/i915_drv.h | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index cf52d4adaa20..b41a414079f4 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -546,7 +546,15 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915,
 #define IS_ALDERLAKE_P(i915) IS_PLATFORM(i915, INTEL_ALDERLAKE_P)
 #define IS_DG2(i915)	IS_PLATFORM(i915, INTEL_DG2)
 #define IS_METEORLAKE(i915) IS_PLATFORM(i915, INTEL_METEORLAKE)
-#define IS_LUNARLAKE(i915) 0
+/*
+ * Display code shared by i915 and Xe relies on macros like IS_LUNARLAKE,
+ * so we need to define these even on platforms that the i915 base driver
+ * doesn't support.  Ensure the parameter is used in the definition to
+ * avoid 'unused variable' warnings when compiling the shared display code
+ * for i915.
+ */
+#define IS_LUNARLAKE(i915) (0 && i915)
+#define IS_BATTLEMAGE(i915)  (0 && i915)
 
 #define IS_DG2_G10(i915) \
 	IS_SUBPLATFORM(i915, INTEL_DG2, INTEL_SUBPLATFORM_G10)
-- 
2.25.1

[PATCH 05/25] drm/i915/xe2: Skip CCS modifiers for Xe2 platforms

[Balasubramani Vivekanandan]

Xe2 platforms doesn't support Aux CCS and the Flat CCS is enabled
through PAT. No CCS modifiers required for Xe2 platforms.

Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_fb.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 3ea6470d6d92..923e97c3aa6c 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -431,9 +431,17 @@ static bool plane_has_modifier(struct drm_i915_private *i915,
 	 * Separate AuxCCS and Flat CCS modifiers to be run only on platforms
 	 * where supported.
 	 */
-	if (intel_fb_is_ccs_modifier(md->modifier) &&
-	    HAS_FLAT_CCS(i915) != !md->ccs.packed_aux_planes)
-		return false;
+	if (intel_fb_is_ccs_modifier(md->modifier)) {
+		/*
+		 * No CCS modifiers available on Xe2 platforms as they don't
+		 * support Aux CCS and the Flat CCS is enabled via PAT
+		 */
+		if ((DISPLAY_VER(i915) >= 20) || IS_BATTLEMAGE(i915))
+			return false;
+
+		if (HAS_FLAT_CCS(i915) != !md->ccs.packed_aux_planes)
+			return false;
+	}
 
 	return true;
 }
-- 
2.25.1

[PATCH 06/25] drm/i915/xe2hpd: Initial cdclk table

[Balasubramani Vivekanandan]

From: Clint Taylor <clinton.a.taylor@intel.com>

Add Xe2_HPD specific CDCLK table and use MTL Funcs.

Bspec: 65243
Cc: Matt Roper <matthew.d.roper@intel.com>
CC: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cdclk.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 31aaa9780dfc..da16c308670f 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -1444,6 +1444,14 @@ static const struct intel_cdclk_vals xe2lpd_cdclk_table[] = {
 	{}
 };
 
+/*
+ * Xe2_HPD always uses the minimal cdclk table from Wa_15015413771
+ */
+static const struct intel_cdclk_vals xe2hpd_cdclk_table[] = {
+	{ .refclk = 38400, .cdclk = 652800, .ratio = 34, .waveform = 0xffff },
+	{}
+};
+
 static const int cdclk_squash_len = 16;
 
 static int cdclk_squash_divider(u16 waveform)
@@ -3768,6 +3776,9 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv)
 	if (DISPLAY_VER(dev_priv) >= 20) {
 		dev_priv->display.funcs.cdclk = &rplu_cdclk_funcs;
 		dev_priv->display.cdclk.table = xe2lpd_cdclk_table;
+	} else if (DISPLAY_VER_FULL(dev_priv) >= IP_VER(14, 1)) {
+		dev_priv->display.funcs.cdclk = &rplu_cdclk_funcs;
+		dev_priv->display.cdclk.table = xe2hpd_cdclk_table;
 	} else if (DISPLAY_VER(dev_priv) >= 14) {
 		dev_priv->display.funcs.cdclk = &rplu_cdclk_funcs;
 		dev_priv->display.cdclk.table = mtl_cdclk_table;
-- 
2.25.1

[PATCH 07/25] Revert "drm/i915/dgfx: DGFX uses direct VBT pin mapping"

[Balasubramani Vivekanandan]

From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>

This reverts commit 562f33836f519a235e5c5e71bcc723ab1faccd2f.
For BMG it seems that the VBT to DDI mapping does not follow DG1, and
DG2, but follows ADLP mapping given in Bspec:20124.

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 2abd2d7ceda2..03fbd6c73f3f 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -2238,15 +2238,14 @@ static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin)
 	const u8 *ddc_pin_map;
 	int i, n_entries;
 
-	if (IS_DGFX(i915))
-		return vbt_pin;
-
 	if (INTEL_PCH_TYPE(i915) >= PCH_MTL || IS_ALDERLAKE_P(i915)) {
 		ddc_pin_map = adlp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(adlp_ddc_pin_map);
 	} else if (IS_ALDERLAKE_S(i915)) {
 		ddc_pin_map = adls_ddc_pin_map;
 		n_entries = ARRAY_SIZE(adls_ddc_pin_map);
+	} else if (INTEL_PCH_TYPE(i915) >= PCH_DG1) {
+		return vbt_pin;
 	} else if (IS_ROCKETLAKE(i915) && INTEL_PCH_TYPE(i915) == PCH_TGP) {
 		ddc_pin_map = rkl_pch_tgp_ddc_pin_map;
 		n_entries = ARRAY_SIZE(rkl_pch_tgp_ddc_pin_map);
-- 
2.25.1

[PATCH 08/25] drm/i915/bmg: Extend DG2 tc check to future

[Balasubramani Vivekanandan]

From: Radhakrishna Sripada <radhakrishna.sripada@intel.com>

Discrete cards use the Port numbers TC1-4 for the offsets. The regular
flow for type-c subsystem port initialization can be skipped. This check
is present in DG2. Extend this to future discrete products.

Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 614e60420a29..aed25890b6f5 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -1861,11 +1861,10 @@ bool intel_phy_is_combo(struct drm_i915_private *dev_priv, enum phy phy)
 bool intel_phy_is_tc(struct drm_i915_private *dev_priv, enum phy phy)
 {
 	/*
-	 * DG2's "TC1", although TC-capable output, doesn't share the same flow
-	 * as other platforms on the display engine side and rather rely on the
-	 * SNPS PHY, that is programmed separately
+	 * Discrete GPU phy's are not attached to FIA's to support TC
+	 * subsystem Legacy or non-legacy, and only support native DP/HDMI
 	 */
-	if (IS_DG2(dev_priv))
+	if (IS_DGFX(dev_priv))
 		return false;
 
 	if (DISPLAY_VER(dev_priv) >= 13)
-- 
2.25.1

[PATCH 09/25] drm/i915/xe2hpd: Properly disable power in port A

[Balasubramani Vivekanandan]

From: José Roberto de Souza <jose.souza@intel.com>

Xe2_HPD has a different value to power down port A.

BSpec: 65450
CC: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 13a2e3db2812..caaae5d3758e 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2921,17 +2921,28 @@ void intel_mtl_pll_enable(struct intel_encoder *encoder,
 		intel_cx0pll_enable(encoder, crtc_state);
 }
 
+static u8 cx0_power_control_disable_val(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	if (intel_encoder_is_c10phy(encoder))
+		return CX0_P2PG_STATE_DISABLE;
+
+	if (IS_BATTLEMAGE(i915) && encoder->port == PORT_A)
+		return CX0_P2PG_STATE_DISABLE;
+
+	return CX0_P4PG_STATE_DISABLE;
+}
+
 static void intel_cx0pll_disable(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
 	enum phy phy = intel_encoder_to_phy(encoder);
-	bool is_c10 = intel_encoder_is_c10phy(encoder);
 	intel_wakeref_t wakeref = intel_cx0_phy_transaction_begin(encoder);
 
 	/* 1. Change owned PHY lane power to Disable state. */
 	intel_cx0_powerdown_change_sequence(encoder, INTEL_CX0_BOTH_LANES,
-					    is_c10 ? CX0_P2PG_STATE_DISABLE :
-					    CX0_P4PG_STATE_DISABLE);
+					    cx0_power_control_disable_val(encoder));
 
 	/*
 	 * 2. Follow the Display Voltage Frequency Switching Sequence Before
-- 
2.25.1

[PATCH 10/25] drm/i915/xe2hpd: Add new C20 PLL register address

[Balasubramani Vivekanandan]

Xe2_HPD has different address for C20 PLL registers. Enable the support
to use the right PLL register address based on display version.

Note that Xe2_LPD uses the same C20 SRAM offsets used by Xe_LPDP (i.e.
MTL's display). According to the BSpec, currently, only Xe2_HPD has
different offsets, so make sure it is the only display using them in the
driver.

Bspec: 67610
Cc: Clint Taylor <Clinton.A.Taylor@intel.com>
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c  | 27 +++++++++++++++++--
 .../gpu/drm/i915/display/intel_cx0_phy_regs.h |  9 +++++++
 2 files changed, 34 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index caaae5d3758e..6e4647859fc6 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -770,6 +770,17 @@ static struct intel_c20pll_reg mtl_c20_reg = {
 	.mpllb_b = MTL_C20_B_MPLLB_CFG_ADDR
 };
 
+static struct intel_c20pll_reg xe2hpd_c20_reg = {
+	.tx_cnt_a = XE2HPD_C20_A_TX_CNTX_CFG_ADDR,
+	.tx_cnt_b = XE2HPD_C20_B_TX_CNTX_CFG_ADDR,
+	.cmn_cnt_a = XE2HPD_C20_A_CMN_CNTX_CFG_ADDR,
+	.cmn_cnt_b = XE2HPD_C20_B_CMN_CNTX_CFG_ADDR,
+	.mplla_a = XE2HPD_C20_A_MPLLA_CFG_ADDR,
+	.mplla_b = XE2HPD_C20_B_MPLLA_CFG_ADDR,
+	.mpllb_a = XE2HPD_C20_A_MPLLB_CFG_ADDR,
+	.mpllb_b = XE2HPD_C20_B_MPLLB_CFG_ADDR,
+};
+
 /* C20 basic DP 1.4 tables */
 static const struct intel_c20pll_state mtl_c20_dp_rbr = {
 	.clock = 162000,
@@ -2166,19 +2177,29 @@ static int intel_c20pll_calc_port_clock(struct intel_encoder *encoder,
 	return vco << tx_rate_mult >> tx_clk_div >> tx_rate;
 }
 
+static struct intel_c20pll_reg *intel_c20_get_pll_reg(struct drm_i915_private *i915)
+{
+	if (DISPLAY_VER_FULL(i915) == IP_VER(14, 1))
+		return &xe2hpd_c20_reg;
+	else
+		return &mtl_c20_reg;
+}
+
 static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder,
 					  struct intel_c20pll_state *pll_state)
 {
 	bool cntx;
 	intel_wakeref_t wakeref;
 	int i;
-	struct intel_c20pll_reg *pll_reg = &mtl_c20_reg;
+	struct intel_c20pll_reg *pll_reg;
 
 	wakeref = intel_cx0_phy_transaction_begin(encoder);
 
 	/* 1. Read current context selection */
 	cntx = intel_cx0_read(encoder, INTEL_CX0_LANE0, PHY_C20_VDR_CUSTOM_SERDES_RATE) & PHY_C20_CONTEXT_TOGGLE;
 
+	pll_reg = intel_c20_get_pll_reg(to_i915(encoder->base.dev));
+
 	/* Read Tx configuration */
 	for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) {
 		if (cntx)
@@ -2353,7 +2374,7 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 	u32 clock = crtc_state->port_clock;
 	bool cntx;
 	int i;
-	const struct intel_c20pll_reg *pll_reg = &mtl_c20_reg;
+	const struct intel_c20pll_reg *pll_reg;
 
 	if (intel_crtc_has_dp_encoder(crtc_state))
 		dp = true;
@@ -2372,6 +2393,8 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 		usleep_range(4000, 4100);
 	}
 
+	pll_reg = intel_c20_get_pll_reg(i915);
+
 	/* 3. Write SRAM configuration context. If A in use, write configuration to B context */
 	/* 3.1 Tx configuration */
 	for (i = 0; i < ARRAY_SIZE(pll_state->tx); i++) {
diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
index 882b98dc347b..8e5fd605b99e 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h
@@ -292,6 +292,15 @@ struct intel_c20pll_reg {
 #define MTL_C20_A_MPLLB_CFG_ADDR	0xCB5A
 #define MTL_C20_B_MPLLB_CFG_ADDR	0xCB4E
 
+#define XE2HPD_C20_A_TX_CNTX_CFG_ADDR	0xCF5E
+#define XE2HPD_C20_B_TX_CNTX_CFG_ADDR	0xCF5A
+#define XE2HPD_C20_A_CMN_CNTX_CFG_ADDR	0xCE8E
+#define XE2HPD_C20_B_CMN_CNTX_CFG_ADDR	0xCE89
+#define XE2HPD_C20_A_MPLLA_CFG_ADDR	0xCE58
+#define XE2HPD_C20_B_MPLLA_CFG_ADDR	0xCE4D
+#define XE2HPD_C20_A_MPLLB_CFG_ADDR	0xCCC2
+#define XE2HPD_C20_B_MPLLB_CFG_ADDR	0xCCB6
+
 /* C20 Phy VSwing Masks */
 #define C20_PHY_VSWING_PREEMPH_MASK	REG_GENMASK8(5, 0)
 #define C20_PHY_VSWING_PREEMPH(val)	REG_FIELD_PREP8(C20_PHY_VSWING_PREEMPH_MASK, val)
-- 
2.25.1

[PATCH 11/25] drm/i915/xe2hpd: Add support for eDP PLL configuration

[Balasubramani Vivekanandan]

Tables for eDP PHY PLL configuration for different link rates added for
Xe2_HPD. Previous platforms were using C10 PHY for eDP port whereas
Xe2_HPD has C20 PHY.

Bpsec: 64568

CC: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 147 ++++++++++++++++++-
 1 file changed, 146 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 6e4647859fc6..d948035f07ad 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -967,6 +967,148 @@ static const struct intel_c20pll_state * const mtl_c20_dp_tables[] = {
 	NULL,
 };
 
+/*
+ * eDP link rates with 38.4 MHz reference clock.
+ */
+
+static const struct intel_c20pll_state xe2hpd_c20_edp_r216 = {
+	.clock = 216000,
+	.tx = { 0xbe88,
+		0x4800,
+		0x0000,
+		},
+	.cmn = { 0x0500,
+		 0x0005,
+		 0x0000,
+		 0x0000,
+		},
+	.mpllb = { 0x50e1,
+		   0x2120,
+		   0x8e18,
+		   0xbfc1,
+		   0x9000,
+		   0x78f6,
+		   0x0000,
+		   0x0000,
+		   0x0000,
+		   0x0000,
+		   0x0000,
+		  },
+};
+
+static const struct intel_c20pll_state xe2hpd_c20_edp_r243 = {
+	.clock = 243000,
+	.tx = { 0xbe88,
+		0x4800,
+		0x0000,
+		},
+	.cmn = { 0x0500,
+		 0x0005,
+		 0x0000,
+		 0x0000,
+		},
+	.mpllb = { 0x50fd,
+		   0x2120,
+		   0x8f18,
+		   0xbfc1,
+		   0xa200,
+		   0x8814,
+		   0x2000,
+		   0x0001,
+		   0x1000,
+		   0x0000,
+		   0x0000,
+		  },
+};
+
+static const struct intel_c20pll_state xe2hpd_c20_edp_r324 = {
+	.clock = 324000,
+	.tx = { 0xbe88,
+		0x4800,
+		0x0000,
+		},
+	.cmn = { 0x0500,
+		 0x0005,
+		 0x0000,
+		 0x0000,
+		},
+	.mpllb = { 0x30a8,
+		   0x2110,
+		   0xcd9a,
+		   0xbfc1,
+		   0x6c00,
+		   0x5ab8,
+		   0x2000,
+		   0x0001,
+		   0x6000,
+		   0x0000,
+		   0x0000,
+		  },
+};
+
+static const struct intel_c20pll_state xe2hpd_c20_edp_r432 = {
+	.clock = 432000,
+	.tx = { 0xbe88,
+		0x4800,
+		0x0000,
+		},
+	.cmn = { 0x0500,
+		 0x0005,
+		 0x0000,
+		 0x0000,
+		},
+	.mpllb = { 0x30e1,
+		   0x2110,
+		   0x8e18,
+		   0xbfc1,
+		   0x9000,
+		   0x78f6,
+		   0x0000,
+		   0x0000,
+		   0x0000,
+		   0x0000,
+		   0x0000,
+		  },
+};
+
+static const struct intel_c20pll_state xe2hpd_c20_edp_r675 = {
+	.clock = 675000,
+	.tx = { 0xbe88,
+		0x4800,
+		0x0000,
+		},
+	.cmn = { 0x0500,
+		 0x0005,
+		 0x0000,
+		 0x0000,
+		},
+	.mpllb = { 0x10af,
+		   0x2108,
+		   0xce1a,
+		   0xbfc1,
+		   0x7080,
+		   0x5e80,
+		   0x2000,
+		   0x0001,
+		   0x6400,
+		   0x0000,
+		   0x0000,
+		  },
+};
+
+static const struct intel_c20pll_state * const xe2hpd_c20_edp_tables[] = {
+	&mtl_c20_dp_rbr,
+	&xe2hpd_c20_edp_r216,
+	&xe2hpd_c20_edp_r243,
+	&mtl_c20_dp_hbr1,
+	&xe2hpd_c20_edp_r324,
+	&xe2hpd_c20_edp_r432,
+	&mtl_c20_dp_hbr2,
+	&xe2hpd_c20_edp_r675,
+	&mtl_c20_dp_hbr3,
+	NULL,
+};
+
 /*
  * HDMI link rates with 38.4 MHz reference clock.
  */
@@ -2084,7 +2226,10 @@ intel_c20_pll_tables_get(struct intel_crtc_state *crtc_state,
 			 struct intel_encoder *encoder)
 {
 	if (intel_crtc_has_dp_encoder(crtc_state))
-		return mtl_c20_dp_tables;
+		if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP))
+			return xe2hpd_c20_edp_tables;
+		else
+			return mtl_c20_dp_tables;
 	else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
 		return mtl_c20_hdmi_tables;
 
-- 
2.25.1

[PATCH 12/25] drm/i915/xe2hpd: update pll values in sync with Bspec

[Balasubramani Vivekanandan]

From: Ravi Kumar Vodapalli <ravi.kumar.vodapalli@intel.com>

DP/eDP and HDMI pll values are updated for Xe2_HPD platform

Bspec: 74165
Signed-off-by: Ravi Kumar Vodapalli <ravi.kumar.vodapalli@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 47 +++++++++++++++++++-
 1 file changed, 45 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index d948035f07ad..20035be015c3 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -1109,6 +1109,42 @@ static const struct intel_c20pll_state * const xe2hpd_c20_edp_tables[] = {
 	NULL,
 };
 
+static const struct intel_c20pll_state xe2hpd_c20_dp_uhbr13_5 = {
+	.clock = 1350000, /* 13.5 Gbps */
+	.tx = {	0xbea0, /* tx cfg0 */
+		0x4800, /* tx cfg1 */
+		0x0000, /* tx cfg2 */
+		},
+	.cmn = {0x0500, /* cmn cfg0*/
+		0x0005, /* cmn cfg1 */
+		0x0000, /* cmn cfg2 */
+		0x0000, /* cmn cfg3 */
+		},
+	.mpllb = { 0x015f,	/* mpllb cfg0 */
+		0x2205,		/* mpllb cfg1 */
+		0x1b17,		/* mpllb cfg2 */
+		0xffc1,		/* mpllb cfg3 */
+		0xbd00,		/* mpllb cfg4 */
+		0x9ec3,		/* mpllb cfg5 */
+		0x2000,		/* mpllb cfg6 */
+		0x0001,		/* mpllb cfg7 */
+		0x4800,		/* mpllb cfg8 */
+		0x0000,		/* mpllb cfg9 */
+		0x0000,		/* mpllb cfg10 */
+		},
+};
+
+static const struct intel_c20pll_state * const xe2hpd_c20_dp_tables[] = {
+	&mtl_c20_dp_rbr,
+	&mtl_c20_dp_hbr1,
+	&mtl_c20_dp_hbr2,
+	&mtl_c20_dp_hbr3,
+	&mtl_c20_dp_uhbr10,
+	&xe2hpd_c20_dp_uhbr13_5,
+	&mtl_c20_dp_uhbr20,
+	NULL,
+};
+
 /*
  * HDMI link rates with 38.4 MHz reference clock.
  */
@@ -2225,13 +2261,20 @@ static const struct intel_c20pll_state * const *
 intel_c20_pll_tables_get(struct intel_crtc_state *crtc_state,
 			 struct intel_encoder *encoder)
 {
-	if (intel_crtc_has_dp_encoder(crtc_state))
+	struct drm_i915_private *i915 = to_i915(encoder->base.dev);
+
+	if (intel_crtc_has_dp_encoder(crtc_state)) {
 		if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP))
 			return xe2hpd_c20_edp_tables;
+
+		if (DISPLAY_VER_FULL(i915) == IP_VER(14, 1))
+			return xe2hpd_c20_dp_tables;
 		else
 			return mtl_c20_dp_tables;
-	else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI))
+
+	} else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) {
 		return mtl_c20_hdmi_tables;
+	}
 
 	MISSING_CASE(encoder->type);
 	return NULL;
-- 
2.25.1

[PATCH 13/25] drm/i915/xe2hpd: Add display info

[Balasubramani Vivekanandan]

From: Lucas De Marchi <lucas.demarchi@intel.com>

Add initial display info for xe2hpd. It is similar to xelpd, but with no
PORT_B.

Bspec: 67066
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 .../gpu/drm/i915/display/intel_display_device.c  | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index c02d79b50006..2c505c480337 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -768,6 +768,21 @@ static const struct intel_display_device_info xe2_lpd_display = {
 		BIT(INTEL_FBC_C) | BIT(INTEL_FBC_D),
 };
 
+static const struct intel_display_device_info xe2_hpd_display = {
+	XE_LPD_FEATURES,
+	.has_cdclk_crawl = 1,
+	.has_cdclk_squash = 1,
+
+	.__runtime_defaults.ip.ver = 14,
+	.__runtime_defaults.ip.rel = 1,
+	.__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B),
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) |
+		BIT(TRANSCODER_C) | BIT(TRANSCODER_D),
+	.__runtime_defaults.port_mask = BIT(PORT_A) |
+		BIT(PORT_TC1) | BIT(PORT_TC2) | BIT(PORT_TC3) | BIT(PORT_TC4),
+};
+
 /*
  * Separate detection for no display cases to keep the display id array simple.
  *
@@ -847,6 +862,7 @@ static const struct {
 	const struct intel_display_device_info *display;
 } gmdid_display_map[] = {
 	{ 14,  0, &xe_lpdp_display },
+	{ 14,  1, &xe2_hpd_display },
 	{ 20,  0, &xe2_lpd_display },
 };
 
-- 
2.25.1

[PATCH 14/25] drm/i915/xe2hpd: Add missing chicken bit register programming

[Balasubramani Vivekanandan]

From: Anusha Srivatsa <anusha.srivatsa@intel.com>

Add step 9 from initialize display sequence.

Bpsec: 49189
Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_power.c | 4 ++++
 drivers/gpu/drm/i915/i915_reg.h                    | 1 +
 2 files changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 6fd4fa52253a..bf9685acf75a 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -1694,6 +1694,10 @@ static void icl_display_core_init(struct drm_i915_private *dev_priv,
 	if (IS_DG2(dev_priv))
 		intel_snps_phy_wait_for_calibration(dev_priv);
 
+	/* 9. XE2_HPD: Program CHICKEN_MISC_2 before any cursor or planes are enabled */
+	if (DISPLAY_VER_FULL(dev_priv) == IP_VER(14, 1))
+		intel_de_rmw(dev_priv, CHICKEN_MISC_2, BMG_DARB_HALF_BLK_END_BURST, 1);
+
 	if (resume)
 		intel_dmc_load_program(dev_priv);
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 58f3e4bfe254..875d76fb8cd0 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4548,6 +4548,7 @@
 
 #define CHICKEN_MISC_2		_MMIO(0x42084)
 #define   CHICKEN_MISC_DISABLE_DPT	REG_BIT(30) /* adl,dg2 */
+#define   BMG_DARB_HALF_BLK_END_BURST	REG_BIT(27)
 #define   KBL_ARB_FILL_SPARE_14		REG_BIT(14)
 #define   KBL_ARB_FILL_SPARE_13		REG_BIT(13)
 #define   GLK_CL2_PWR_DOWN		REG_BIT(12)
-- 
2.25.1

[PATCH 15/25] drm/xe/display: Lane reversal requires writes to both context lanes

[Balasubramani Vivekanandan]

From: Clint Taylor <clinton.a.taylor@intel.com>

Write both CX0 Lanes for Context Toggle for all except TC pin assignment D.

BSPEC: 64539
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 20035be015c3..cbcb6651dfed 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -2558,7 +2558,7 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 {
 	const struct intel_c20pll_state *pll_state = &crtc_state->cx0pll_state.c20;
 	bool dp = false;
-	int lane = crtc_state->lane_count > 2 ? INTEL_CX0_BOTH_LANES : INTEL_CX0_LANE0;
+	u8 owned_lane_mask = intel_cx0_get_owned_lane_mask(encoder);
 	u32 clock = crtc_state->port_clock;
 	bool cntx;
 	int i;
@@ -2634,19 +2634,19 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 	}
 
 	/* 4. Program custom width to match the link protocol */
-	intel_cx0_rmw(encoder, lane, PHY_C20_VDR_CUSTOM_WIDTH,
+	intel_cx0_rmw(encoder, owned_lane_mask, PHY_C20_VDR_CUSTOM_WIDTH,
 		      PHY_C20_CUSTOM_WIDTH_MASK,
 		      PHY_C20_CUSTOM_WIDTH(intel_get_c20_custom_width(clock, dp)),
 		      MB_WRITE_COMMITTED);
 
 	/* 5. For DP or 6. For HDMI */
 	if (dp) {
-		intel_cx0_rmw(encoder, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE,
+		intel_cx0_rmw(encoder, owned_lane_mask, PHY_C20_VDR_CUSTOM_SERDES_RATE,
 			      BIT(6) | PHY_C20_CUSTOM_SERDES_MASK,
 			      BIT(6) | PHY_C20_CUSTOM_SERDES(intel_c20_get_dp_rate(clock)),
 			      MB_WRITE_COMMITTED);
 	} else {
-		intel_cx0_rmw(encoder, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE,
+		intel_cx0_rmw(encoder, owned_lane_mask, PHY_C20_VDR_CUSTOM_SERDES_RATE,
 			      BIT(7) | PHY_C20_CUSTOM_SERDES_MASK,
 			      is_hdmi_frl(clock) ? BIT(7) : 0,
 			      MB_WRITE_COMMITTED);
@@ -2660,7 +2660,7 @@ static void intel_c20_pll_program(struct drm_i915_private *i915,
 	 * 7. Write Vendor specific registers to toggle context setting to load
 	 * the updated programming toggle context bit
 	 */
-	intel_cx0_rmw(encoder, lane, PHY_C20_VDR_CUSTOM_SERDES_RATE,
+	intel_cx0_rmw(encoder, owned_lane_mask, PHY_C20_VDR_CUSTOM_SERDES_RATE,
 		      BIT(0), cntx ? 0 : 1, MB_WRITE_COMMITTED);
 }
 
-- 
2.25.1

[PATCH 16/25] drm/xe/xe2hpd: Define a new DRAM type INTEL_DRAM_GDDR

[Balasubramani Vivekanandan]

Defined a new DRAM type to be used in the following patches.
The following patch first makes use of this new type in the i915
display. So without this define, build would fail when the shared
display code is built for Xe.

Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/xe/xe_device_types.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 1df3dcc17d75..e7aa2dd3df8d 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -480,6 +480,7 @@ struct xe_device {
 			INTEL_DRAM_LPDDR4,
 			INTEL_DRAM_DDR5,
 			INTEL_DRAM_LPDDR5,
+			INTEL_DRAM_GDDR,
 		} type;
 		u8 num_qgv_points;
 		u8 num_psf_gv_points;
-- 
2.25.1

[PATCH 17/25] drm/i915/xe2hpd: Add max memory bandwidth algorithm

[Balasubramani Vivekanandan]

From: Matt Roper <matthew.d.roper@intel.com>

Unlike DG2, Xe2_HPD does support multiple GV points with different
maximum memory bandwidths, but uses a much simpler algorithm than igpu
platforms use.

Bspec: 64631
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bw.c | 65 ++++++++++++++++++++++++-
 drivers/gpu/drm/i915/i915_drv.h         |  1 +
 drivers/gpu/drm/i915/soc/intel_dram.c   |  4 ++
 3 files changed, 68 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 7f2a50b4f494..dc9ac4831065 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -22,6 +22,8 @@ struct intel_qgv_point {
 	u16 dclk, t_rp, t_rdpre, t_rc, t_ras, t_rcd;
 };
 
+#define DEPROGBWPCLIMIT		60
+
 struct intel_psf_gv_point {
 	u8 clk; /* clock in multiples of 16.6666 MHz */
 };
@@ -239,6 +241,9 @@ static int icl_get_qgv_points(struct drm_i915_private *dev_priv,
 			qi->channel_width = 16;
 			qi->deinterleave = 4;
 			break;
+		case INTEL_DRAM_GDDR:
+			qi->channel_width = 32;
+			break;
 		default:
 			MISSING_CASE(dram_info->type);
 			return -EINVAL;
@@ -383,6 +388,12 @@ static const struct intel_sa_info mtl_sa_info = {
 	.derating = 10,
 };
 
+static const struct intel_sa_info xe2_hpd_sa_info = {
+	.derating = 30,
+	.deprogbwlimit = 53,
+	/* Other values not used by simplified algorithm */
+};
+
 static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa)
 {
 	struct intel_qgv_info qi = {};
@@ -489,7 +500,7 @@ static int tgl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel
 	dclk_max = icl_sagv_max_dclk(&qi);
 
 	peakbw = num_channels * DIV_ROUND_UP(qi.channel_width, 8) * dclk_max;
-	maxdebw = min(sa->deprogbwlimit * 1000, peakbw * 6 / 10); /* 60% */
+	maxdebw = min(sa->deprogbwlimit * 1000, peakbw * DEPROGBWPCLIMIT / 100);
 
 	ipqdepth = min(ipqdepthpch, sa->displayrtids / num_channels);
 	/*
@@ -594,6 +605,54 @@ static void dg2_get_bw_info(struct drm_i915_private *i915)
 	i915->display.sagv.status = I915_SAGV_NOT_CONTROLLED;
 }
 
+static int xe2_hpd_get_bw_info(struct drm_i915_private *i915,
+			       const struct intel_sa_info *sa)
+{
+	struct intel_qgv_info qi = {};
+	int num_channels = i915->dram_info.num_channels;
+	int peakbw, maxdebw;
+	int ret, i;
+
+	ret = icl_get_qgv_points(i915, &qi, true);
+	if (ret) {
+		drm_dbg_kms(&i915->drm,
+			    "Failed to get memory subsystem information, ignoring bandwidth limits");
+		return ret;
+	}
+
+	peakbw = num_channels * qi.channel_width / 8 * icl_sagv_max_dclk(&qi);
+	maxdebw = min(sa->deprogbwlimit * 1000, peakbw * DEPROGBWPCLIMIT / 10);
+
+	for (i = 0; i < qi.num_points; i++) {
+		const struct intel_qgv_point *point = &qi.points[i];
+		int bw = num_channels * (qi.channel_width / 8) * point->dclk;
+
+		i915->display.bw.max[0].deratedbw[i] =
+			min(maxdebw, (100 - sa->derating) * bw / 100);
+		i915->display.bw.max[0].peakbw[i] = bw;
+
+		drm_dbg_kms(&i915->drm, "QGV %d: deratedbw=%u peakbw: %u\n",
+			    i, i915->display.bw.max[0].deratedbw[i],
+			    i915->display.bw.max[0].peakbw[i]);
+	}
+
+	/* Bandwidth does not depend on # of planes; set all groups the same */
+	i915->display.bw.max[0].num_planes = 1;
+	i915->display.bw.max[0].num_qgv_points = qi.num_points;
+	for (i = 1; i < ARRAY_SIZE(i915->display.bw.max); i++)
+		memcpy(&i915->display.bw.max[i], &i915->display.bw.max[0],
+		       sizeof(i915->display.bw.max[0]));
+
+	/*
+	 * Xe2_HPD should always have exactly two QGV points representing
+	 * battery and plugged-in operation.
+	 */
+	drm_WARN_ON(&i915->drm, qi.num_points != 2);
+	i915->display.sagv.status = I915_SAGV_ENABLED;
+
+	return 0;
+}
+
 static unsigned int icl_max_bw_index(struct drm_i915_private *dev_priv,
 				     int num_planes, int qgv_point)
 {
@@ -664,7 +723,9 @@ void intel_bw_init_hw(struct drm_i915_private *dev_priv)
 	if (!HAS_DISPLAY(dev_priv))
 		return;
 
-	if (DISPLAY_VER(dev_priv) >= 14)
+	if (DISPLAY_VER_FULL(dev_priv) >= IP_VER(14, 1) && IS_DGFX(dev_priv))
+		xe2_hpd_get_bw_info(dev_priv, &xe2_hpd_sa_info);
+	else if (DISPLAY_VER(dev_priv) >= 14)
 		tgl_get_bw_info(dev_priv, &mtl_sa_info);
 	else if (IS_DG2(dev_priv))
 		dg2_get_bw_info(dev_priv);
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index b41a414079f4..eb3ad6ae0b2a 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -313,6 +313,7 @@ struct drm_i915_private {
 			INTEL_DRAM_LPDDR4,
 			INTEL_DRAM_DDR5,
 			INTEL_DRAM_LPDDR5,
+			INTEL_DRAM_GDDR,
 		} type;
 		u8 num_qgv_points;
 		u8 num_psf_gv_points;
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 15492b69f698..99b541babb31 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -640,6 +640,10 @@ static int xelpdp_get_dram_info(struct drm_i915_private *i915)
 	case 5:
 		dram_info->type = INTEL_DRAM_LPDDR3;
 		break;
+	case 8:
+		drm_WARN_ON(&i915->drm, !IS_DGFX(i915));
+		dram_info->type = INTEL_DRAM_GDDR;
+		break;
 	default:
 		MISSING_CASE(val);
 		return -EINVAL;
-- 
2.25.1

[PATCH 18/25] drm/i915/display: Enable RM timeout detection

[Balasubramani Vivekanandan]

From: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>

Enable RM timeout interrupt to detect any hang during display engine
register access. This interrupt is supported only on Display version 14.
Current default timeout is 2ms.

WA: 14012195489
Bspec: 50110

CC: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_irq.c | 10 ++++++++++
 drivers/gpu/drm/i915/i915_reg.h                  |  3 +++
 2 files changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index f846c5b108b5..3035b50fcad9 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -851,6 +851,13 @@ gen8_de_misc_irq_handler(struct drm_i915_private *dev_priv, u32 iir)
 {
 	bool found = false;
 
+	if (iir & GEN8_DE_RM_TIMEOUT) {
+		u32 val = intel_uncore_read(&dev_priv->uncore,
+				RMTIMEOUTREG_CAPTURE);
+		drm_warn(&dev_priv->drm, "Register Access Timeout = 0x%x\n", val);
+		found = true;
+	}
+
 	if (DISPLAY_VER(dev_priv) >= 14) {
 		if (iir & (XELPDP_PMDEMAND_RSP |
 			   XELPDP_PMDEMAND_RSPTOUT_ERR)) {
@@ -1666,6 +1673,9 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 			de_port_masked |= DSI0_TE | DSI1_TE;
 	}
 
+	if (DISPLAY_VER(dev_priv) == 14)
+		de_misc_masked |= GEN8_DE_RM_TIMEOUT;
+
 	de_pipe_enables = de_pipe_masked |
 		GEN8_PIPE_VBLANK |
 		gen8_de_pipe_underrun_mask(dev_priv) |
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 875d76fb8cd0..d1692b32bb8a 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4212,6 +4212,8 @@
 #define RM_TIMEOUT		_MMIO(0x42060)
 #define  MMIO_TIMEOUT_US(us)	((us) << 0)
 
+#define RMTIMEOUTREG_CAPTURE	_MMIO(0x420e0)
+
 /* interrupts */
 #define DE_MASTER_IRQ_CONTROL   (1 << 31)
 #define DE_SPRITEB_FLIP_DONE    (1 << 29)
@@ -4398,6 +4400,7 @@
 #define GEN8_DE_MISC_IMR _MMIO(0x44464)
 #define GEN8_DE_MISC_IIR _MMIO(0x44468)
 #define GEN8_DE_MISC_IER _MMIO(0x4446c)
+#define  GEN8_DE_RM_TIMEOUT		REG_BIT(29)
 #define  XELPDP_PMDEMAND_RSPTOUT_ERR	REG_BIT(27)
 #define  GEN8_DE_MISC_GSE		REG_BIT(27)
 #define  GEN8_DE_EDP_PSR		REG_BIT(19)
-- 
2.25.1

[PATCH 19/25] drm/i915/xe2hpd: Do not program MBUS_DBOX BW credits

[Balasubramani Vivekanandan]

From: José Roberto de Souza <jose.souza@intel.com>

Xe2_HPD doesn't have DBOX BW credits, so here programing it with
zero.

BSpec: 49213
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/skl_watermark.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index bc341abcab2f..22ae782e89f4 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -3733,7 +3733,7 @@ void intel_mbus_dbox_update(struct intel_atomic_state *state)
 		if (!new_crtc_state->hw.active)
 			continue;
 
-		if (DISPLAY_VER(i915) >= 14) {
+		if (DISPLAY_VER(i915) >= 14 && !IS_BATTLEMAGE(i915)) {
 			if (xelpdp_is_only_pipe_per_dbuf_bank(crtc->pipe,
 							      new_dbuf_state->active_pipes))
 				pipe_val |= MBUS_DBOX_BW_8CREDITS_MTL;
-- 
2.25.1

[PATCH 20/25] drm/i915/bmg: BMG should re-use MTL's south display logic

[Balasubramani Vivekanandan]

From: Matt Roper <matthew.d.roper@intel.com>

Battlemage's south display is the same as Meteor Lake's, including the
need to invert the HPD pins, which Lunar Lake does not need.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_pch.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_pch.c b/drivers/gpu/drm/i915/soc/intel_pch.c
index 3cad6dac06b0..542eea50093c 100644
--- a/drivers/gpu/drm/i915/soc/intel_pch.c
+++ b/drivers/gpu/drm/i915/soc/intel_pch.c
@@ -218,10 +218,10 @@ void intel_detect_pch(struct drm_i915_private *dev_priv)
 	if (DISPLAY_VER(dev_priv) >= 20) {
 		dev_priv->pch_type = PCH_LNL;
 		return;
-	} else if (IS_METEORLAKE(dev_priv)) {
+	} else if (IS_BATTLEMAGE(dev_priv) || IS_METEORLAKE(dev_priv)) {
 		/*
 		 * Both north display and south display are on the SoC die.
-		 * The real PCH is uninvolved in display.
+		 * The real PCH (if it even exists) is uninvolved in display.
 		 */
 		dev_priv->pch_type = PCH_MTL;
 		return;
-- 
2.25.1

[PATCH 21/25] drm/i915/xe2hpd: Set maximum DP rate to UHBR13.5

[Balasubramani Vivekanandan]

Max supported speed by xe2hpd is UHBR13.5. Limit the max DP source rate
to it.

Bspec: 67066

Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index b393ddbb7b35..d9d37f4971dd 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -466,6 +466,9 @@ static int mtl_max_source_rate(struct intel_dp *intel_dp)
 	if (intel_encoder_is_c10phy(encoder))
 		return 810000;
 
+	if (DISPLAY_VER_FULL(to_i915(encoder->base.dev)) == IP_VER(14, 1))
+		return 1350000;
+
 	return 2000000;
 }
 
-- 
2.25.1

[PATCH 22/25] drm/xe/gt_print: add xe_gt_err_once()

[Balasubramani Vivekanandan]

From: Matthew Auld <matthew.auld@intel.com>

Needed in an upcoming patch, where we want GT level print, but only
which to trigger once to avoid flooding dmesg.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_printk.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_gt_printk.h b/drivers/gpu/drm/xe/xe_gt_printk.h
index c2b004d3f48e..d6228baaff1e 100644
--- a/drivers/gpu/drm/xe/xe_gt_printk.h
+++ b/drivers/gpu/drm/xe/xe_gt_printk.h
@@ -13,6 +13,9 @@
 #define xe_gt_printk(_gt, _level, _fmt, ...) \
 	drm_##_level(&gt_to_xe(_gt)->drm, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
 
+#define xe_gt_err_once(_gt, _fmt, ...) \
+	xe_gt_printk((_gt), err_once, _fmt, ##__VA_ARGS__)
+
 #define xe_gt_err(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), err, _fmt, ##__VA_ARGS__)
 
-- 
2.25.1

[PATCH 23/25] drm/xe/device: implement transient flush

[Balasubramani Vivekanandan]

From: Nirmoy Das <nirmoy.das@intel.com>

Display surfaces can be tagged as transient by mapping it using one of
the various L3:XD PAT index modes on Xe2. The expectation is that KMD
needs to request transient data flush at the start of flip sequence to
ensure all transient data in L3 cache is flushed to memory. Add a
routine for this which we can then call from the display code.

Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Co-developed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_gt_regs.h |  3 ++
 drivers/gpu/drm/xe/xe_device.c       | 49 ++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_device.h       |  2 ++
 3 files changed, 54 insertions(+)

diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
index d5b21f03beaa..9c6549830e24 100644
--- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
@@ -305,6 +305,9 @@
 
 #define XE2LPM_L3SQCREG5			XE_REG_MCR(0xb658)
 
+#define XE2_TDF_CTRL				XE_REG(0xb418)
+#define   TRANSIENT_FLUSH_REQUEST		REG_BIT(0)
+
 #define XEHP_MERT_MOD_CTRL			XE_REG_MCR(0xcf28)
 #define RENDER_MOD_CTRL				XE_REG_MCR(0xcf2c)
 #define COMP_MOD_CTRL				XE_REG_MCR(0xcf30)
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 01bd5ccf05ca..66182220e663 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -641,6 +641,55 @@ void xe_device_wmb(struct xe_device *xe)
 		xe_mmio_write32(gt, SOFTWARE_FLAGS_SPR33, 0);
 }
 
+/**
+ * xe_device_td_flush() - Flush transient L3 cache entries
+ * @xe: The device
+ *
+ * Display engine has direct access to memory and is never coherent with L3/L4
+ * caches (or CPU caches), however KMD is responsible for specifically flushing
+ * transient L3 GPU cache entries prior to the flip sequence to ensure scanout
+ * can happen from such a surface without seeing corruption.
+ *
+ * Display surfaces can be tagged as transient by mapping it using one of the
+ * various L3:XD PAT index modes on Xe2.
+ *
+ * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed
+ * at the end of each submission via PIPE_CONTROL for compute/render, since SA
+ * Media is not coherent with L3 and we want to support render-vs-media
+ * usescases. For other engines like copy/blt the HW internally forces uncached
+ * behaviour, hence why we can skip the TDF on such platforms.
+ */
+void xe_device_td_flush(struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	u8 id;
+
+	if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20)
+		return;
+
+	for_each_gt(gt, xe, id) {
+		if (xe_gt_is_media_type(gt))
+			continue;
+
+		xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+
+		xe_mmio_write32(gt, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST);
+		/*
+		 * FIXME: We can likely do better here with our choice of
+		 * timeout.  Currently we just assume the worst case, but really
+		 * we should make this dependent on how much actual L3 there is
+		 * for this system. Recomendation is to allow ~64us in the worst
+		 * case for 8M of L3 (assumes all entries are transient and need
+		 * to be flushed).
+		 */
+		if (xe_mmio_wait32(gt, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST, 0,
+				   150, NULL, false))
+			xe_gt_err_once(gt, "TD flush timeout\n");
+
+		xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
+	}
+}
+
 u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size)
 {
 	return xe_device_has_flat_ccs(xe) ?
diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
index d413bc2c6be5..d3430f4b820a 100644
--- a/drivers/gpu/drm/xe/xe_device.h
+++ b/drivers/gpu/drm/xe/xe_device.h
@@ -176,4 +176,6 @@ void xe_device_snapshot_print(struct xe_device *xe, struct drm_printer *p);
 u64 xe_device_canonicalize_addr(struct xe_device *xe, u64 address);
 u64 xe_device_uncanonicalize_addr(struct xe_device *xe, u64 address);
 
+void xe_device_td_flush(struct xe_device *xe);
+
 #endif
-- 
2.25.1

[PATCH 24/25] drm/i915/display: perform transient flush

[Balasubramani Vivekanandan]

From: Matthew Auld <matthew.auld@intel.com>

Perform manual transient cache flush prior to flip and at the end of
frontbuffer_flush. This is needed to ensure display engine doesn't see
garbage if the surface is L3:XD dirty.

Testcase: igt@xe-pat@display-vs-wb-transient
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c  |  3 +++
 .../gpu/drm/i915/display/intel_frontbuffer.c  |  2 ++
 drivers/gpu/drm/i915/display/intel_tdf.h      | 25 +++++++++++++++++++
 drivers/gpu/drm/xe/Makefile                   |  3 ++-
 drivers/gpu/drm/xe/display/xe_tdf.c           | 13 ++++++++++
 5 files changed, 45 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_tdf.h
 create mode 100644 drivers/gpu/drm/xe/display/xe_tdf.c

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index aed25890b6f5..0a720e9d12a7 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -110,6 +110,7 @@
 #include "intel_sdvo.h"
 #include "intel_snps_phy.h"
 #include "intel_tc.h"
+#include "intel_tdf.h"
 #include "intel_tv.h"
 #include "intel_vblank.h"
 #include "intel_vdsc.h"
@@ -7095,6 +7096,8 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state)
 
 	intel_atomic_commit_fence_wait(state);
 
+	intel_td_flush(dev_priv);
+
 	drm_atomic_helper_wait_for_dependencies(&state->base);
 	drm_dp_mst_atomic_wait_for_dependencies(&state->base);
 	intel_atomic_global_state_wait_for_dependencies(state);
diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.c b/drivers/gpu/drm/i915/display/intel_frontbuffer.c
index 2ea37c0414a9..4923c340a0b6 100644
--- a/drivers/gpu/drm/i915/display/intel_frontbuffer.c
+++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.c
@@ -65,6 +65,7 @@
 #include "intel_fbc.h"
 #include "intel_frontbuffer.h"
 #include "intel_psr.h"
+#include "intel_tdf.h"
 
 /**
  * frontbuffer_flush - flush frontbuffer
@@ -93,6 +94,7 @@ static void frontbuffer_flush(struct drm_i915_private *i915,
 	trace_intel_frontbuffer_flush(i915, frontbuffer_bits, origin);
 
 	might_sleep();
+	intel_td_flush(i915);
 	intel_drrs_flush(i915, frontbuffer_bits);
 	intel_psr_flush(i915, frontbuffer_bits, origin);
 	intel_fbc_flush(i915, frontbuffer_bits, origin);
diff --git a/drivers/gpu/drm/i915/display/intel_tdf.h b/drivers/gpu/drm/i915/display/intel_tdf.h
new file mode 100644
index 000000000000..353cde21f6c2
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_tdf.h
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2024 Intel Corporation
+ */
+
+#ifndef __INTEL_TDF_H__
+#define __INTEL_TDF_H__
+
+/*
+ * TDF (Transient-Data-Flush) is needed for Xe2+ where special L3:XD caching can
+ * be enabled through various PAT index modes. Idea is to use this caching mode
+ * when for example rendering onto the display surface, with the promise that
+ * KMD will ensure transient cache entries are always flushed by the time we do
+ * the display flip, since display engine is never coherent with CPU/GPU caches.
+ */
+
+struct drm_i915_private;
+
+#ifdef I915
+static inline void intel_td_flush(struct drm_i915_private *i915) {}
+#else
+void intel_td_flush(struct drm_i915_private *i915);
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 21316ee47026..71847e33f4c2 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -199,7 +199,8 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	display/xe_dsb_buffer.o \
 	display/xe_fb_pin.o \
 	display/xe_hdcp_gsc.o \
-	display/xe_plane_initial.o
+	display/xe_plane_initial.o \
+	display/xe_tdf.o
 
 # SOC code shared with i915
 xe-$(CONFIG_DRM_XE_DISPLAY) += \
diff --git a/drivers/gpu/drm/xe/display/xe_tdf.c b/drivers/gpu/drm/xe/display/xe_tdf.c
new file mode 100644
index 000000000000..2c0d4e144e09
--- /dev/null
+++ b/drivers/gpu/drm/xe/display/xe_tdf.c
@@ -0,0 +1,13 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2024 Intel Corporation
+ */
+
+#include "xe_device.h"
+#include "intel_display_types.h"
+#include "intel_tdf.h"
+
+void intel_td_flush(struct drm_i915_private *i915)
+{
+	xe_device_td_flush(i915);
+}
-- 
2.25.1

[PATCH 25/25] drm/xe/bmg: Enable the display support

[Balasubramani Vivekanandan]

Enable the display support for Battlemage

Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
---
 drivers/gpu/drm/xe/xe_pci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index b3158053baee..835c18ec8fb9 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -340,6 +340,7 @@ static const struct xe_device_desc lnl_desc = {
 static const struct xe_device_desc bmg_desc = {
 	DGFX_FEATURES,
 	PLATFORM(XE_BATTLEMAGE),
+	.has_display = true,
 	.require_force_probe = true,
 };
 
-- 
2.25.1

✗ CI.Patch_applied: failure for Enable dislay support for Battlemage

[Patchwork]

== Series Details ==

Series: Enable dislay support for Battlemage
URL   : https://patchwork.freedesktop.org/series/131983/
State : failure

== Summary ==

=== Applying kernel patches on branch 'drm-tip' with base: ===
Base commit: 23e6199ddb93 drm-tip: 2024y-04m-03d-07h-14m-06s UTC integration manifest
=== git am output follows ===
error: patch failed: drivers/gpu/drm/i915/display/intel_display_device.c:768
error: drivers/gpu/drm/i915/display/intel_display_device.c: patch does not apply
hint: Use 'git am --show-current-patch=diff' to see the failed patch
Applying: drm/i915/display: Prepare to handle new C20 PLL register address
Applying: drm/xe/bmg: Add BMG platform definition
Applying: drm/xe/bmg: Define IS_BATTLEMAGE macro
Applying: drm/i915/bmg: Define IS_BATTLEMAGE macro
Applying: drm/i915/xe2: Skip CCS modifiers for Xe2 platforms
Applying: drm/i915/xe2hpd: Initial cdclk table
Applying: Revert "drm/i915/dgfx: DGFX uses direct VBT pin mapping"
Applying: drm/i915/bmg: Extend DG2 tc check to future
Applying: drm/i915/xe2hpd: Properly disable power in port A
Applying: drm/i915/xe2hpd: Add new C20 PLL register address
Applying: drm/i915/xe2hpd: Add support for eDP PLL configuration
Applying: drm/i915/xe2hpd: update pll values in sync with Bspec
Applying: drm/i915/xe2hpd: Add display info
Patch failed at 0013 drm/i915/xe2hpd: Add display info
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.
To restore the original branch and stop patching, run "git am --abort".

Re: [PATCH 23/25] drm/xe/device: implement transient flush

[Nirmoy Das]

There is new fixup patch(PR#630) which modifies this patch. Could you 
please bring that in as well.


Regards,

Nirmoy

On 4/3/2024 12:51 PM, Balasubramani Vivekanandan wrote:
> From: Nirmoy Das <nirmoy.das@intel.com>
>
> Display surfaces can be tagged as transient by mapping it using one of
> the various L3:XD PAT index modes on Xe2. The expectation is that KMD
> needs to request transient data flush at the start of flip sequence to
> ensure all transient data in L3 cache is flushed to memory. Add a
> routine for this which we can then call from the display code.
>
> Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
> Co-developed-by: Matthew Auld <matthew.auld@intel.com>
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
> ---
>   drivers/gpu/drm/xe/regs/xe_gt_regs.h |  3 ++
>   drivers/gpu/drm/xe/xe_device.c       | 49 ++++++++++++++++++++++++++++
>   drivers/gpu/drm/xe/xe_device.h       |  2 ++
>   3 files changed, 54 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> index d5b21f03beaa..9c6549830e24 100644
> --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
> @@ -305,6 +305,9 @@
>   
>   #define XE2LPM_L3SQCREG5			XE_REG_MCR(0xb658)
>   
> +#define XE2_TDF_CTRL				XE_REG(0xb418)
> +#define   TRANSIENT_FLUSH_REQUEST		REG_BIT(0)
> +
>   #define XEHP_MERT_MOD_CTRL			XE_REG_MCR(0xcf28)
>   #define RENDER_MOD_CTRL				XE_REG_MCR(0xcf2c)
>   #define COMP_MOD_CTRL				XE_REG_MCR(0xcf30)
> diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
> index 01bd5ccf05ca..66182220e663 100644
> --- a/drivers/gpu/drm/xe/xe_device.c
> +++ b/drivers/gpu/drm/xe/xe_device.c
> @@ -641,6 +641,55 @@ void xe_device_wmb(struct xe_device *xe)
>   		xe_mmio_write32(gt, SOFTWARE_FLAGS_SPR33, 0);
>   }
>   
> +/**
> + * xe_device_td_flush() - Flush transient L3 cache entries
> + * @xe: The device
> + *
> + * Display engine has direct access to memory and is never coherent with L3/L4
> + * caches (or CPU caches), however KMD is responsible for specifically flushing
> + * transient L3 GPU cache entries prior to the flip sequence to ensure scanout
> + * can happen from such a surface without seeing corruption.
> + *
> + * Display surfaces can be tagged as transient by mapping it using one of the
> + * various L3:XD PAT index modes on Xe2.
> + *
> + * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed
> + * at the end of each submission via PIPE_CONTROL for compute/render, since SA
> + * Media is not coherent with L3 and we want to support render-vs-media
> + * usescases. For other engines like copy/blt the HW internally forces uncached
> + * behaviour, hence why we can skip the TDF on such platforms.
> + */
> +void xe_device_td_flush(struct xe_device *xe)
> +{
> +	struct xe_gt *gt;
> +	u8 id;
> +
> +	if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20)
> +		return;
> +
> +	for_each_gt(gt, xe, id) {
> +		if (xe_gt_is_media_type(gt))
> +			continue;
> +
> +		xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
> +
> +		xe_mmio_write32(gt, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST);
> +		/*
> +		 * FIXME: We can likely do better here with our choice of
> +		 * timeout.  Currently we just assume the worst case, but really
> +		 * we should make this dependent on how much actual L3 there is
> +		 * for this system. Recomendation is to allow ~64us in the worst
> +		 * case for 8M of L3 (assumes all entries are transient and need
> +		 * to be flushed).
> +		 */
> +		if (xe_mmio_wait32(gt, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST, 0,
> +				   150, NULL, false))
> +			xe_gt_err_once(gt, "TD flush timeout\n");
> +
> +		xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
> +	}
> +}
> +
>   u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size)
>   {
>   	return xe_device_has_flat_ccs(xe) ?
> diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
> index d413bc2c6be5..d3430f4b820a 100644
> --- a/drivers/gpu/drm/xe/xe_device.h
> +++ b/drivers/gpu/drm/xe/xe_device.h
> @@ -176,4 +176,6 @@ void xe_device_snapshot_print(struct xe_device *xe, struct drm_printer *p);
>   u64 xe_device_canonicalize_addr(struct xe_device *xe, u64 address);
>   u64 xe_device_uncanonicalize_addr(struct xe_device *xe, u64 address);
>   
> +void xe_device_td_flush(struct xe_device *xe);
> +
>   #endif