drm/xe/ptl: Apply Wa_14023061436

Enable WMTP for the BTD kernel to address Wa14023061436 by setting the
proper TDL Chicken Bit.

v2: Apply it on engine_was[] as this register is not part of LRC(Matt)
    Apply it for first_render_or_compute in case this gets extended to
    compute only platforms(Matt).

Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250108141323.311601-1-nirmoy.das@intel.com
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>

Author: nirmoy

drivers/gpu/drm/xe/regs/xe_gt_regs.h

@@ -500,6 +500,9 @@
 #define   LSC_L1_FLUSH_CTL_3D_DATAPORT_FLUSH_EVENTS_MASK	REG_GENMASK(13, 11)
 #define   DIS_ATOMIC_CHAINING_TYPED_WRITES	REG_BIT(3)
 
+#define TDL_CHICKEN				XE_REG_MCR(0xe5f4, XE_REG_OPTION_MASKED)
+#define   QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE	REG_BIT(12)
+
 #define LSC_CHICKEN_BIT_0			XE_REG_MCR(0xe7c8)
 #define   DISABLE_D8_D16_COASLESCE		REG_BIT(30)
 #define   WR_REQ_CHAINING_DIS			REG_BIT(26)

drivers/gpu/drm/xe/xe_wa.c

@@ -613,6 +613,11 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	  XE_RTP_ACTIONS(FIELD_SET(SAMPLER_MODE, SMP_WAIT_FETCH_MERGING_COUNTER,
 				   SMP_FORCE_128B_OVERFETCH))
 	},
+	{ XE_RTP_NAME("14023061436"),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001),
+		       FUNC(xe_rtp_match_first_render_or_compute)),
+	  XE_RTP_ACTIONS(SET(TDL_CHICKEN, QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE))
+	},
 
 	{}
 };