[Debug-WIP]drm/xe: enable ptl sriov vf driver (WA)

drivers/gpu/drm/xe/Makefile

@@ -34,8 +34,8 @@ xe-y += xe_bb.o \
 	xe_dma_buf.o \
 	xe_drm_client.o \
 	xe_exec.o \
-	xe_execlist.o \
 	xe_exec_queue.o \
+	xe_execlist.o \
 	xe_force_wake.o \
 	xe_ggtt.o \
 	xe_gpu_scheduler.o \
@@ -56,6 +56,7 @@ xe-y += xe_bb.o \
 	xe_gt_topology.o \
 	xe_guc.o \
 	xe_guc_ads.o \
+	xe_guc_buf.o \
 	xe_guc_capture.o \
 	xe_guc_ct.o \
 	xe_guc_db_mgr.o \
@@ -66,11 +67,11 @@ xe-y += xe_bb.o \
 	xe_guc_pc.o \
 	xe_guc_submit.o \
 	xe_heci_gsc.o \
+	xe_huc.o \
 	xe_hw_engine.o \
 	xe_hw_engine_class_sysfs.o \
 	xe_hw_engine_group.o \
 	xe_hw_fence.o \
-	xe_huc.o \
 	xe_irq.o \
 	xe_lrc.o \
 	xe_migrate.o \
@@ -90,36 +91,40 @@ xe-y += xe_bb.o \
 	xe_range_fence.o \
 	xe_reg_sr.o \
 	xe_reg_whitelist.o \
-	xe_rtp.o \
 	xe_ring_ops.o \
+	xe_rtp.o \
 	xe_sa.o \
 	xe_sched_job.o \
 	xe_step.o \
+	xe_survivability_mode.o \
 	xe_sync.o \
 	xe_tile.o \
 	xe_tile_sysfs.o \
 	xe_trace.o \
 	xe_trace_bo.o \
 	xe_trace_guc.o \
 	xe_trace_lrc.o \
-	xe_ttm_sys_mgr.o \
 	xe_ttm_stolen_mgr.o \
+	xe_ttm_sys_mgr.o \
 	xe_ttm_vram_mgr.o \
 	xe_tuning.o \
 	xe_uc.o \
 	xe_uc_fw.o \
 	xe_vm.o \
 	xe_vram.o \
 	xe_vram_freq.o \
-	xe_wait_user_fence.o \
+	xe_vsec.o \
 	xe_wa.o \
+	xe_wait_user_fence.o \
 	xe_wopcm.o
 
 xe-$(CONFIG_HMM_MIRROR) += xe_hmm.o
 
 # graphics hardware monitoring (HWMON) support
 xe-$(CONFIG_HWMON) += xe_hwmon.o
 
+xe-$(CONFIG_PERF_EVENTS) += xe_pmu.o
+
 # graphics virtualization (SR-IOV) support
 xe-y += \
 	xe_gt_sriov_vf.o \
@@ -219,6 +224,7 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	i915-display/intel_display_wa.o \
 	i915-display/intel_dkl_phy.o \
 	i915-display/intel_dmc.o \
+	i915-display/intel_dmc_wl.o \
 	i915-display/intel_dp.o \
 	i915-display/intel_dp_aux.o \
 	i915-display/intel_dp_aux_backlight.o \
@@ -266,7 +272,6 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	i915-display/intel_vdsc.o \
 	i915-display/intel_vga.o \
 	i915-display/intel_vrr.o \
-	i915-display/intel_dmc_wl.o \
 	i915-display/intel_wm.o \
 	i915-display/skl_scaler.o \
 	i915-display/skl_universal_plane.o \

drivers/gpu/drm/xe/abi/guc_capture_abi.h

@@ -25,7 +25,7 @@ enum guc_state_capture_type {
 
 #define GUC_STATE_CAPTURE_TYPE_MAX	(GUC_STATE_CAPTURE_TYPE_ENGINE_INSTANCE + 1)
 
-/* Class indecies for capture_class and capture_instance arrays */
+/* Class indices for capture_class and capture_instance arrays */
 enum guc_capture_list_class_type {
 	GUC_CAPTURE_LIST_CLASS_RENDER_COMPUTE = 0,
 	GUC_CAPTURE_LIST_CLASS_VIDEO = 1,

drivers/gpu/drm/xe/abi/guc_klvs_abi.h

@@ -132,7 +132,7 @@ enum  {
  * _`GUC_KLV_VGT_POLICY_SCHED_IF_IDLE` : 0x8001
  *      This config sets whether strict scheduling is enabled whereby any VF
  *      that doesn’t have work to submit is still allocated a fixed execution
- *      time-slice to ensure active VFs execution is always consitent even
+ *      time-slice to ensure active VFs execution is always consistent even
  *      during other VF reprovisiong / rebooting events. Changing this KLV
  *      impacts all VFs and takes effect on the next VF-Switch event.
  *
@@ -207,7 +207,7 @@ enum  {
  *      of and this will never be perfectly-exact (accumulated nano-second
  *      granularity) since the GPUs clock time runs off a different crystal
  *      from the CPUs clock. Changing this KLV on a VF that is currently
- *      running a context wont take effect until a new context is scheduled in.
+ *      running a context won't take effect until a new context is scheduled in.
  *      That said, when the PF is changing this value from 0x0 to
  *      a non-zero value, it might never take effect if the VF is running an
  *      infinitely long compute or shader kernel. In such a scenario, the
@@ -227,7 +227,7 @@ enum  {
  *      HW is capable and this will never be perfectly-exact (accumulated
  *      nano-second granularity) since the GPUs clock time runs off a
  *      different crystal from the CPUs clock. Changing this KLV on a VF
- *      that is currently running a context wont take effect until a new
+ *      that is currently running a context won't take effect until a new
  *      context is scheduled in.
  *      That said, when the PF is changing this value from 0x0 to
  *      a non-zero value, it might never take effect if the VF is running an

drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h

@@ -1,3 +1,8 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2024 Intel Corporation
+ */
+
 #ifndef _I915_GEM_STOLEN_H_
 #define _I915_GEM_STOLEN_H_
 

drivers/gpu/drm/xe/compat-i915-headers/intel_uncore.h

@@ -117,19 +117,10 @@ __intel_wait_for_register(struct intel_uncore *uncore, i915_reg_t i915_reg,
 			  unsigned int slow_timeout_ms, u32 *out_value)
 {
 	struct xe_reg reg = XE_REG(i915_mmio_reg_offset(i915_reg));
-	bool atomic;
-
-	/*
-	 * Replicate the behavior from i915 here, in which sleep is not
-	 * performed if slow_timeout_ms == 0. This is necessary because
-	 * of some paths in display code where waits are done in atomic
-	 * context.
-	 */
-	atomic = !slow_timeout_ms && fast_timeout_us > 0;
 
 	return xe_mmio_wait32(__compat_uncore_to_mmio(uncore), reg, mask, value,
 			      fast_timeout_us + 1000 * slow_timeout_ms,
-			      out_value, atomic);
+			      out_value, false);
 }
 
 static inline u32 intel_uncore_read_fw(struct intel_uncore *uncore,

drivers/gpu/drm/xe/display/ext/i915_irq.c

@@ -64,7 +64,7 @@ bool intel_irqs_enabled(struct xe_device *xe)
 	 * But at this point the xe irq is better protected against races,
 	 * although the full solution would be protecting the i915 side.
 	 */
-	return xe->irq.enabled;
+	return atomic_read(&xe->irq.enabled);
 }
 
 void intel_synchronize_irq(struct xe_device *xe)

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

@@ -83,6 +83,8 @@
 #define RING_IMR(base)				XE_REG((base) + 0xa8)
 #define RING_INT_STATUS_RPT_PTR(base)		XE_REG((base) + 0xac)
 
+#define CS_INT_VEC(base)			XE_REG((base) + 0x1b8)
+
 #define RING_EIR(base)				XE_REG((base) + 0xb0)
 #define RING_EMR(base)				XE_REG((base) + 0xb4)
 #define RING_ESR(base)				XE_REG((base) + 0xb8)
@@ -138,6 +140,7 @@
 
 #define RING_MODE(base)				XE_REG((base) + 0x29c)
 #define   GFX_DISABLE_LEGACY_MODE		REG_BIT(3)
+#define   GFX_MSIX_INTERRUPT_ENABLE		REG_BIT(13)
 
 #define RING_TIMESTAMP(base)			XE_REG((base) + 0x358)
 

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

@@ -221,6 +221,9 @@
 
 #define MIRROR_FUSE1				XE_REG(0x911c)
 
+#define MIRROR_L3BANK_ENABLE			XE_REG(0x9130)
+#define   XE3_L3BANK_ENABLE			REG_GENMASK(31, 0)
+
 #define XELP_EU_ENABLE				XE_REG(0x9134)	/* "_DISABLE" on Xe_LP */
 #define   XELP_EU_MASK				REG_GENMASK(7, 0)
 #define XELP_GT_SLICE_ENABLE			XE_REG(0x9138)
@@ -410,12 +413,6 @@
 
 #define XE2LPM_SCRATCH3_LBCF			XE_REG_MCR(0xb654)
 
-#define XE2LPM_L3SQCREG2			XE_REG_MCR(0xb604)
-
-#define XE2LPM_L3SQCREG3			XE_REG_MCR(0xb608)
-
-#define XE2LPM_SCRATCH3_LBCF			XE_REG_MCR(0xb654)
-
 #define XE2LPM_L3SQCREG5			XE_REG_MCR(0xb658)
 
 #define XE2_TDF_CTRL				XE_REG(0xb418)
@@ -506,6 +503,9 @@
 #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/regs/xe_lrc_layout.h

@@ -25,6 +25,9 @@
 #define CTX_INT_SRC_REPORT_REG		(CTX_LRI_INT_REPORT_PTR + 3)
 #define CTX_INT_SRC_REPORT_PTR		(CTX_LRI_INT_REPORT_PTR + 4)
 
+#define CTX_CS_INT_VEC_REG		0x5a
+#define CTX_CS_INT_VEC_DATA		(CTX_CS_INT_VEC_REG + 1)
+
 #define INDIRECT_CTX_RING_HEAD		(0x02 + 1)
 #define INDIRECT_CTX_RING_TAIL		(0x04 + 1)
 #define INDIRECT_CTX_RING_START		(0x06 + 1)

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

@@ -41,14 +41,6 @@
 
 #define OAG_OABUFFER		XE_REG(0xdb08)
 #define  OABUFFER_SIZE_MASK	REG_GENMASK(5, 3)
-#define  OABUFFER_SIZE_128K	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 0)
-#define  OABUFFER_SIZE_256K	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 1)
-#define  OABUFFER_SIZE_512K	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 2)
-#define  OABUFFER_SIZE_1M	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 3)
-#define  OABUFFER_SIZE_2M	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 4)
-#define  OABUFFER_SIZE_4M	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 5)
-#define  OABUFFER_SIZE_8M	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 6)
-#define  OABUFFER_SIZE_16M	REG_FIELD_PREP(OABUFFER_SIZE_MASK, 7)
 #define  OAG_OABUFFER_MEMORY_SELECT		REG_BIT(0) /* 0: PPGTT, 1: GGTT */
 
 #define OAG_OACONTROL				XE_REG(0xdaf4)
@@ -59,10 +51,15 @@
 /* Common to all OA units */
 #define  OA_OACONTROL_REPORT_BC_MASK		REG_GENMASK(9, 9)
 #define  OA_OACONTROL_COUNTER_SIZE_MASK		REG_GENMASK(8, 8)
+#define  OAG_OACONTROL_USED_BITS \
+	(OAG_OACONTROL_OA_PES_DISAG_EN | OAG_OACONTROL_OA_CCS_SELECT_MASK | \
+	 OAG_OACONTROL_OA_COUNTER_SEL_MASK | OAG_OACONTROL_OA_COUNTER_ENABLE | \
+	 OA_OACONTROL_REPORT_BC_MASK | OA_OACONTROL_COUNTER_SIZE_MASK)
 
 #define OAG_OA_DEBUG XE_REG(0xdaf8, XE_REG_OPTION_MASKED)
 #define  OAG_OA_DEBUG_DISABLE_MMIO_TRG			REG_BIT(14)
 #define  OAG_OA_DEBUG_START_TRIGGER_SCOPE_CONTROL	REG_BIT(13)
+#define  OAG_OA_DEBUG_BUF_SIZE_SELECT			REG_BIT(12)
 #define  OAG_OA_DEBUG_DISABLE_START_TRG_2_COUNT_QUAL	REG_BIT(8)
 #define  OAG_OA_DEBUG_DISABLE_START_TRG_1_COUNT_QUAL	REG_BIT(7)
 #define  OAG_OA_DEBUG_INCLUDE_CLK_RATIO			REG_BIT(6)
@@ -85,6 +82,8 @@
 #define OAM_CONTEXT_CONTROL_OFFSET		(0x1bc)
 #define OAM_CONTROL_OFFSET			(0x194)
 #define  OAM_CONTROL_COUNTER_SEL_MASK		REG_GENMASK(3, 1)
+#define  OAM_OACONTROL_USED_BITS \
+	(OAM_CONTROL_COUNTER_SEL_MASK | OAG_OACONTROL_OA_COUNTER_ENABLE)
 #define OAM_DEBUG_OFFSET			(0x198)
 #define OAM_STATUS_OFFSET			(0x19c)
 #define OAM_MMIO_TRG_OFFSET			(0x1d0)

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

@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2024 Intel Corporation
+ */
+#ifndef _XE_PMT_H_
+#define _XE_PMT_H_
+
+#define SOC_BASE			0x280000
+
+#define BMG_PMT_BASE_OFFSET		0xDB000
+#define BMG_DISCOVERY_OFFSET		(SOC_BASE + BMG_PMT_BASE_OFFSET)
+
+#define BMG_TELEMETRY_BASE_OFFSET	0xE0000
+#define BMG_TELEMETRY_OFFSET		(SOC_BASE + BMG_TELEMETRY_BASE_OFFSET)
+
+#define SG_REMAP_INDEX1			XE_REG(SOC_BASE + 0x08)
+#define   SG_REMAP_BITS			REG_GENMASK(31, 24)
+
+#endif

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

@@ -13,15 +13,15 @@
 /**
  * struct xe_reg - Register definition
  *
- * Register defintion to be used by the individual register. Although the same
+ * Register definition to be used by the individual register. Although the same
  * definition is used for xe_reg and xe_reg_mcr, they use different internal
  * APIs for accesses.
  */
 struct xe_reg {
 	union {
 		struct {
 			/** @addr: address */
-			u32 addr:28;
+			u32 addr:22;
 			/**
 			 * @masked: register is "masked", with upper 16bits used
 			 * to identify the bits that are updated on the lower
@@ -41,10 +41,6 @@ struct xe_reg {
 			 * @vf: register is accessible from the Virtual Function.
 			 */
 			u32 vf:1;
-			/**
-			 * @ext: access MMIO extension space for current register.
-			 */
-			u32 ext:1;
 		};
 		/** @raw: Raw value with both address and options */
 		u32 raw;
@@ -111,16 +107,6 @@ struct xe_reg_mcr {
  */
 #define XE_REG(r_, ...)		((const struct xe_reg)XE_REG_INITIALIZER(r_, ##__VA_ARGS__))
 
-/**
- * XE_REG_EXT - Create a struct xe_reg from extension offset and additional
- * flags
- * @r_: Register extension offset
- * @...: Additional options like access mode. See struct xe_reg for available
- *       options.
- */
-#define XE_REG_EXT(r_, ...)	\
-	((const struct xe_reg)XE_REG_INITIALIZER(r_, ##__VA_ARGS__, .ext = 1))
-
 /**
  * XE_REG_MCR - Create a struct xe_reg_mcr from offset and additional flags
  * @r_: Register offset

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

@@ -44,12 +44,16 @@
 
 #define MTL_RP_STATE_CAP			XE_REG(0x138000)
 
+#define MTL_GT_RPA_FREQUENCY			XE_REG(0x138008)
 #define MTL_GT_RPE_FREQUENCY			XE_REG(0x13800c)
 
 #define MTL_MEDIAP_STATE_CAP			XE_REG(0x138020)
 #define   MTL_RPN_CAP_MASK			REG_GENMASK(24, 16)
 #define   MTL_RP0_CAP_MASK			REG_GENMASK(8, 0)
 
+#define MTL_MPA_FREQUENCY			XE_REG(0x138028)
+#define   MTL_RPA_MASK				REG_GENMASK(8, 0)
+
 #define MTL_MPE_FREQUENCY			XE_REG(0x13802c)
 #define   MTL_RPE_MASK				REG_GENMASK(8, 0)
 

drivers/gpu/drm/xe/tests/xe_bo.c

@@ -8,7 +8,7 @@
 
 #include <linux/iosys-map.h>
 #include <linux/math64.h>
-#include <linux/random.h>
+#include <linux/prandom.h>
 #include <linux/swap.h>
 
 #include <uapi/linux/sysinfo.h>
@@ -264,10 +264,9 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 		 * however seems quite fragile not to also restart the GT. Try
 		 * to do that here by triggering a GT reset.
 		 */
-		for_each_gt(__gt, xe, id) {
-			xe_gt_reset_async(__gt);
-			flush_work(&__gt->reset.worker);
-		}
+		for_each_gt(__gt, xe, id)
+			xe_gt_reset(__gt);
+
 		if (err) {
 			KUNIT_FAIL(test, "restore kernel err=%pe\n",
 				   ERR_PTR(err));
@@ -606,8 +605,6 @@ static void xe_bo_shrink_kunit(struct kunit *test)
 static struct kunit_case xe_bo_tests[] = {
 	KUNIT_CASE_PARAM(xe_ccs_migrate_kunit, xe_pci_live_device_gen_param),
 	KUNIT_CASE_PARAM(xe_bo_evict_kunit, xe_pci_live_device_gen_param),
-	KUNIT_CASE_PARAM_ATTR(xe_bo_shrink_kunit, xe_pci_live_device_gen_param,
-			      {.speed = KUNIT_SPEED_SLOW}),
 	{}
 };
 
@@ -618,3 +615,17 @@ struct kunit_suite xe_bo_test_suite = {
 	.init = xe_kunit_helper_xe_device_live_test_init,
 };
 EXPORT_SYMBOL_IF_KUNIT(xe_bo_test_suite);
+
+static struct kunit_case xe_bo_shrink_test[] = {
+	KUNIT_CASE_PARAM_ATTR(xe_bo_shrink_kunit, xe_pci_live_device_gen_param,
+			      {.speed = KUNIT_SPEED_SLOW}),
+	{}
+};
+
+VISIBLE_IF_KUNIT
+struct kunit_suite xe_bo_shrink_test_suite = {
+	.name = "xe_bo_shrink",
+	.test_cases = xe_bo_shrink_test,
+	.init = xe_kunit_helper_xe_device_live_test_init,
+};
+EXPORT_SYMBOL_IF_KUNIT(xe_bo_shrink_test_suite);