Merge tag 'drm-xe-fixes-2024-08-22' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes

UAPI Changes:
- Fix OA format masks which were breaking build with gcc-5 (Geert)

Driver Changes:
- Fix opregion leak (Lucas)
- Fix OA sysfs entry (Ashutosh)
- Fix VM dma-resv lock (Brost)
- Fix tile fini sequence (Brost)
- Prevent UAF around preempt fence (Auld)
- Fix DGFX display suspend/resume (Maarten)
- Many Xe/Xe2 critical workarounds (Auld, Ngai-Mint, Bommu, Tejas, Daniele)
- Fix devm/drmm issues (Daniele)
- Fix missing workqueue destroy in xe_gt_pagefault (Stuart)
- Drop HW fence pointer to HW fence ctx (Brost)
- Free job before xe_exec_queue_put (Brost)
Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/ZsdVe0XI2Pq8C-ON@intel.com

drivers/gpu/drm/xe/Makefile

@@ -25,12 +25,14 @@ $(obj)/generated/%_wa_oob.c $(obj)/generated/%_wa_oob.h: $(obj)/xe_gen_wa_oob \
 
 uses_generated_oob := \
 	$(obj)/xe_ggtt.o \
+	$(obj)/xe_device.o \
 	$(obj)/xe_gsc.o \
 	$(obj)/xe_gt.o \
 	$(obj)/xe_guc.o \
 	$(obj)/xe_guc_ads.o \
 	$(obj)/xe_guc_pc.o \
 	$(obj)/xe_migrate.o \
+	$(obj)/xe_pat.o \
 	$(obj)/xe_ring_ops.o \
 	$(obj)/xe_vm.o \
 	$(obj)/xe_wa.o \

drivers/gpu/drm/xe/display/xe_display.c

@@ -132,6 +132,7 @@ static void xe_display_fini_noirq(void *arg)
 		return;
 
 	intel_display_driver_remove_noirq(xe);
+	intel_opregion_cleanup(xe);
 }
 
 int xe_display_init_noirq(struct xe_device *xe)
@@ -157,8 +158,10 @@ int xe_display_init_noirq(struct xe_device *xe)
 	intel_display_device_info_runtime_init(xe);
 
 	err = intel_display_driver_probe_noirq(xe);
-	if (err)
+	if (err) {
+		intel_opregion_cleanup(xe);
 		return err;
+	}
 
 	return devm_add_action_or_reset(xe->drm.dev, xe_display_fini_noirq, xe);
 }
@@ -280,6 +283,27 @@ static bool suspend_to_idle(void)
 	return false;
 }
 
+static void xe_display_flush_cleanup_work(struct xe_device *xe)
+{
+	struct intel_crtc *crtc;
+
+	for_each_intel_crtc(&xe->drm, crtc) {
+		struct drm_crtc_commit *commit;
+
+		spin_lock(&crtc->base.commit_lock);
+		commit = list_first_entry_or_null(&crtc->base.commit_list,
+						  struct drm_crtc_commit, commit_entry);
+		if (commit)
+			drm_crtc_commit_get(commit);
+		spin_unlock(&crtc->base.commit_lock);
+
+		if (commit) {
+			wait_for_completion(&commit->cleanup_done);
+			drm_crtc_commit_put(commit);
+		}
+	}
+}
+
 void xe_display_pm_suspend(struct xe_device *xe, bool runtime)
 {
 	bool s2idle = suspend_to_idle();
@@ -297,6 +321,8 @@ void xe_display_pm_suspend(struct xe_device *xe, bool runtime)
 	if (!runtime)
 		intel_display_driver_suspend(xe);
 
+	xe_display_flush_cleanup_work(xe);
+
 	intel_dp_mst_suspend(xe);
 
 	intel_hpd_cancel_work(xe);

drivers/gpu/drm/xe/display/xe_dsb_buffer.c

@@ -7,6 +7,8 @@
 #include "intel_display_types.h"
 #include "intel_dsb_buffer.h"
 #include "xe_bo.h"
+#include "xe_device.h"
+#include "xe_device_types.h"
 #include "xe_gt.h"
 
 u32 intel_dsb_buffer_ggtt_offset(struct intel_dsb_buffer *dsb_buf)
@@ -16,7 +18,10 @@ u32 intel_dsb_buffer_ggtt_offset(struct intel_dsb_buffer *dsb_buf)
 
 void intel_dsb_buffer_write(struct intel_dsb_buffer *dsb_buf, u32 idx, u32 val)
 {
+	struct xe_device *xe = dsb_buf->vma->bo->tile->xe;
+
 	iosys_map_wr(&dsb_buf->vma->bo->vmap, idx * 4, u32, val);
+	xe_device_l2_flush(xe);
 }
 
 u32 intel_dsb_buffer_read(struct intel_dsb_buffer *dsb_buf, u32 idx)
@@ -26,9 +31,12 @@ u32 intel_dsb_buffer_read(struct intel_dsb_buffer *dsb_buf, u32 idx)
 
 void intel_dsb_buffer_memset(struct intel_dsb_buffer *dsb_buf, u32 idx, u32 val, size_t size)
 {
+	struct xe_device *xe = dsb_buf->vma->bo->tile->xe;
+
 	WARN_ON(idx > (dsb_buf->buf_size - size) / sizeof(*dsb_buf->cmd_buf));
 
 	iosys_map_memset(&dsb_buf->vma->bo->vmap, idx * 4, val, size);
+	xe_device_l2_flush(xe);
 }
 
 bool intel_dsb_buffer_create(struct intel_crtc *crtc, struct intel_dsb_buffer *dsb_buf, size_t size)

drivers/gpu/drm/xe/display/xe_fb_pin.c

@@ -10,6 +10,7 @@
 #include "intel_fb.h"
 #include "intel_fb_pin.h"
 #include "xe_bo.h"
+#include "xe_device.h"
 #include "xe_ggtt.h"
 #include "xe_gt.h"
 #include "xe_pm.h"
@@ -304,6 +305,8 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 	if (ret)
 		goto err_unpin;
 
+	/* Ensure DPT writes are flushed */
+	xe_device_l2_flush(xe);
 	return vma;
 
 err_unpin:

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

@@ -80,6 +80,9 @@
 #define   LE_CACHEABILITY_MASK			REG_GENMASK(1, 0)
 #define   LE_CACHEABILITY(value)		REG_FIELD_PREP(LE_CACHEABILITY_MASK, value)
 
+#define XE2_GAMREQSTRM_CTRL			XE_REG(0x4194)
+#define   CG_DIS_CNTLBUS			REG_BIT(6)
+
 #define CCS_AUX_INV				XE_REG(0x4208)
 
 #define VD0_AUX_INV				XE_REG(0x4218)
@@ -372,6 +375,11 @@
 
 #define XEHPC_L3CLOS_MASK(i)			XE_REG_MCR(0xb194 + (i) * 8)
 
+#define XE2_GLOBAL_INVAL			XE_REG(0xb404)
+
+#define SCRATCH1LPFC				XE_REG(0xb474)
+#define   EN_L3_RW_CCS_CACHE_FLUSH		REG_BIT(0)
+
 #define XE2LPM_L3SQCREG5			XE_REG_MCR(0xb658)
 
 #define XE2_TDF_CTRL				XE_REG(0xb418)
@@ -429,6 +437,7 @@
 #define   DIS_FIX_EOT1_FLUSH			REG_BIT(9)
 
 #define TDL_TSL_CHICKEN				XE_REG_MCR(0xe4c4, XE_REG_OPTION_MASKED)
+#define   STK_ID_RESTRICT			REG_BIT(12)
 #define   SLM_WMTP_RESTORE			REG_BIT(11)
 
 #define ROW_CHICKEN				XE_REG_MCR(0xe4f0, XE_REG_OPTION_MASKED)

drivers/gpu/drm/xe/xe_bo.c

@@ -1575,7 +1575,7 @@ struct xe_bo *xe_bo_create_from_data(struct xe_device *xe, struct xe_tile *tile,
 	return bo;
 }
 
-static void __xe_bo_unpin_map_no_vm(struct drm_device *drm, void *arg)
+static void __xe_bo_unpin_map_no_vm(void *arg)
 {
 	xe_bo_unpin_map_no_vm(arg);
 }
@@ -1590,7 +1590,7 @@ struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile
 	if (IS_ERR(bo))
 		return bo;
 
-	ret = drmm_add_action_or_reset(&xe->drm, __xe_bo_unpin_map_no_vm, bo);
+	ret = devm_add_action_or_reset(xe->drm.dev, __xe_bo_unpin_map_no_vm, bo);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -1638,7 +1638,7 @@ int xe_managed_bo_reinit_in_vram(struct xe_device *xe, struct xe_tile *tile, str
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
-	drmm_release_action(&xe->drm, __xe_bo_unpin_map_no_vm, *src);
+	devm_release_action(xe->drm.dev, __xe_bo_unpin_map_no_vm, *src);
 	*src = bo;
 
 	return 0;

drivers/gpu/drm/xe/xe_device.c

@@ -54,6 +54,9 @@
 #include "xe_vm.h"
 #include "xe_vram.h"
 #include "xe_wait_user_fence.h"
+#include "xe_wa.h"
+
+#include <generated/xe_wa_oob.h>
 
 static int xe_file_open(struct drm_device *dev, struct drm_file *file)
 {
@@ -820,6 +823,11 @@ void xe_device_td_flush(struct xe_device *xe)
 	if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20)
 		return;
 
+	if (XE_WA(xe_root_mmio_gt(xe), 16023588340)) {
+		xe_device_l2_flush(xe);
+		return;
+	}
+
 	for_each_gt(gt, xe, id) {
 		if (xe_gt_is_media_type(gt))
 			continue;
@@ -843,6 +851,30 @@ void xe_device_td_flush(struct xe_device *xe)
 	}
 }
 
+void xe_device_l2_flush(struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	int err;
+
+	gt = xe_root_mmio_gt(xe);
+
+	if (!XE_WA(gt, 16023588340))
+		return;
+
+	err = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+	if (err)
+		return;
+
+	spin_lock(&gt->global_invl_lock);
+	xe_mmio_write32(gt, XE2_GLOBAL_INVAL, 0x1);
+
+	if (xe_mmio_wait32(gt, XE2_GLOBAL_INVAL, 0x1, 0x0, 150, NULL, true))
+		xe_gt_err_once(gt, "Global invalidation timeout\n");
+	spin_unlock(&gt->global_invl_lock);
+
+	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) ?

drivers/gpu/drm/xe/xe_device.h

@@ -162,6 +162,7 @@ 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);
+void xe_device_l2_flush(struct xe_device *xe);
 
 static inline bool xe_device_wedged(struct xe_device *xe)
 {

drivers/gpu/drm/xe/xe_exec_queue.c

@@ -105,22 +105,35 @@ static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe,
 
 static int __xe_exec_queue_init(struct xe_exec_queue *q)
 {
+	struct xe_vm *vm = q->vm;
 	int i, err;
 
+	if (vm) {
+		err = xe_vm_lock(vm, true);
+		if (err)
+			return err;
+	}
+
 	for (i = 0; i < q->width; ++i) {
 		q->lrc[i] = xe_lrc_create(q->hwe, q->vm, SZ_16K);
 		if (IS_ERR(q->lrc[i])) {
 			err = PTR_ERR(q->lrc[i]);
-			goto err_lrc;
+			goto err_unlock;
 		}
 	}
 
+	if (vm)
+		xe_vm_unlock(vm);
+
 	err = q->ops->init(q);
 	if (err)
 		goto err_lrc;
 
 	return 0;
 
+err_unlock:
+	if (vm)
+		xe_vm_unlock(vm);
 err_lrc:
 	for (i = i - 1; i >= 0; --i)
 		xe_lrc_put(q->lrc[i]);
@@ -140,15 +153,7 @@ struct xe_exec_queue *xe_exec_queue_create(struct xe_device *xe, struct xe_vm *v
 	if (IS_ERR(q))
 		return q;
 
-	if (vm) {
-		err = xe_vm_lock(vm, true);
-		if (err)
-			goto err_post_alloc;
-	}
-
 	err = __xe_exec_queue_init(q);
-	if (vm)
-		xe_vm_unlock(vm);
 	if (err)
 		goto err_post_alloc;
 
@@ -638,7 +643,6 @@ int xe_exec_queue_create_ioctl(struct drm_device *dev, void *data,
 
 		if (xe_vm_in_preempt_fence_mode(vm)) {
 			q->lr.context = dma_fence_context_alloc(1);
-			spin_lock_init(&q->lr.lock);
 
 			err = xe_vm_add_compute_exec_queue(vm, q);
 			if (XE_IOCTL_DBG(xe, err))

drivers/gpu/drm/xe/xe_exec_queue_types.h

@@ -126,8 +126,6 @@ struct xe_exec_queue {
 		u32 seqno;
 		/** @lr.link: link into VM's list of exec queues */
 		struct list_head link;
-		/** @lr.lock: preemption fences lock */
-		spinlock_t lock;
 	} lr;
 
 	/** @ops: submission backend exec queue operations */

drivers/gpu/drm/xe/xe_gsc.c

@@ -260,7 +260,7 @@ static int gsc_upload_and_init(struct xe_gsc *gsc)
 	struct xe_tile *tile = gt_to_tile(gt);
 	int ret;
 
-	if (XE_WA(gt, 14018094691)) {
+	if (XE_WA(tile->primary_gt, 14018094691)) {
 		ret = xe_force_wake_get(gt_to_fw(tile->primary_gt), XE_FORCEWAKE_ALL);
 
 		/*
@@ -278,7 +278,7 @@ static int gsc_upload_and_init(struct xe_gsc *gsc)
 
 	ret = gsc_upload(gsc);
 
-	if (XE_WA(gt, 14018094691))
+	if (XE_WA(tile->primary_gt, 14018094691))
 		xe_force_wake_put(gt_to_fw(tile->primary_gt), XE_FORCEWAKE_ALL);
 
 	if (ret)
@@ -437,7 +437,7 @@ int xe_gsc_init(struct xe_gsc *gsc)
 	return ret;
 }
 
-static void free_resources(struct drm_device *drm, void *arg)
+static void free_resources(void *arg)
 {
 	struct xe_gsc *gsc = arg;
 
@@ -501,7 +501,7 @@ int xe_gsc_init_post_hwconfig(struct xe_gsc *gsc)
 	gsc->q = q;
 	gsc->wq = wq;
 
-	err = drmm_add_action_or_reset(&xe->drm, free_resources, gsc);
+	err = devm_add_action_or_reset(xe->drm.dev, free_resources, gsc);
 	if (err)
 		return err;
 

drivers/gpu/drm/xe/xe_gt.c

@@ -11,6 +11,8 @@
 #include <drm/xe_drm.h>
 #include <generated/xe_wa_oob.h>
 
+#include <generated/xe_wa_oob.h>
+
 #include "instructions/xe_gfxpipe_commands.h"
 #include "instructions/xe_mi_commands.h"
 #include "regs/xe_gt_regs.h"
@@ -95,6 +97,51 @@ void xe_gt_sanitize(struct xe_gt *gt)
 	gt->uc.guc.submission_state.enabled = false;
 }
 
+static void xe_gt_enable_host_l2_vram(struct xe_gt *gt)
+{
+	u32 reg;
+	int err;
+
+	if (!XE_WA(gt, 16023588340))
+		return;
+
+	err = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+	if (WARN_ON(err))
+		return;
+
+	if (!xe_gt_is_media_type(gt)) {
+		xe_mmio_write32(gt, SCRATCH1LPFC, EN_L3_RW_CCS_CACHE_FLUSH);
+		reg = xe_mmio_read32(gt, XE2_GAMREQSTRM_CTRL);
+		reg |= CG_DIS_CNTLBUS;
+		xe_mmio_write32(gt, XE2_GAMREQSTRM_CTRL, reg);
+	}
+
+	xe_gt_mcr_multicast_write(gt, XEHPC_L3CLOS_MASK(3), 0x3);
+	xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
+}
+
+static void xe_gt_disable_host_l2_vram(struct xe_gt *gt)
+{
+	u32 reg;
+	int err;
+
+	if (!XE_WA(gt, 16023588340))
+		return;
+
+	if (xe_gt_is_media_type(gt))
+		return;
+
+	err = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+	if (WARN_ON(err))
+		return;
+
+	reg = xe_mmio_read32(gt, XE2_GAMREQSTRM_CTRL);
+	reg &= ~CG_DIS_CNTLBUS;
+	xe_mmio_write32(gt, XE2_GAMREQSTRM_CTRL, reg);
+
+	xe_force_wake_put(gt_to_fw(gt), XE_FW_GT);
+}
+
 /**
  * xe_gt_remove() - Clean up the GT structures before driver removal
  * @gt: the GT object
@@ -111,6 +158,8 @@ void xe_gt_remove(struct xe_gt *gt)
 
 	for (i = 0; i < XE_ENGINE_CLASS_MAX; ++i)
 		xe_hw_fence_irq_finish(&gt->fence_irq[i]);
+
+	xe_gt_disable_host_l2_vram(gt);
 }
 
 static void gt_reset_worker(struct work_struct *w);
@@ -339,6 +388,7 @@ int xe_gt_init_early(struct xe_gt *gt)
 
 	xe_force_wake_init_gt(gt, gt_to_fw(gt));
 	xe_pcode_init(gt);
+	spin_lock_init(&gt->global_invl_lock);
 
 	return 0;
 }
@@ -508,6 +558,7 @@ int xe_gt_init_hwconfig(struct xe_gt *gt)
 
 	xe_gt_mcr_init_early(gt);
 	xe_pat_init(gt);
+	xe_gt_enable_host_l2_vram(gt);
 
 	err = xe_uc_init(&gt->uc);
 	if (err)
@@ -643,6 +694,8 @@ static int do_gt_restart(struct xe_gt *gt)
 
 	xe_pat_init(gt);
 
+	xe_gt_enable_host_l2_vram(gt);
+
 	xe_gt_mcr_set_implicit_defaults(gt);
 	xe_reg_sr_apply_mmio(&gt->reg_sr, gt);
 
@@ -796,6 +849,8 @@ int xe_gt_suspend(struct xe_gt *gt)
 
 	xe_gt_idle_disable_pg(gt);
 
+	xe_gt_disable_host_l2_vram(gt);
+
 	XE_WARN_ON(xe_force_wake_put(gt_to_fw(gt), XE_FORCEWAKE_ALL));
 	xe_gt_dbg(gt, "suspended\n");
 

drivers/gpu/drm/xe/xe_gt_pagefault.c

@@ -382,6 +382,18 @@ static void pf_queue_work_func(struct work_struct *w)
 
 static void acc_queue_work_func(struct work_struct *w);
 
+static void pagefault_fini(void *arg)
+{
+	struct xe_gt *gt = arg;
+	struct xe_device *xe = gt_to_xe(gt);
+
+	if (!xe->info.has_usm)
+		return;
+
+	destroy_workqueue(gt->usm.acc_wq);
+	destroy_workqueue(gt->usm.pf_wq);
+}
+
 int xe_gt_pagefault_init(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
@@ -409,10 +421,12 @@ int xe_gt_pagefault_init(struct xe_gt *gt)
 	gt->usm.acc_wq = alloc_workqueue("xe_gt_access_counter_work_queue",
 					 WQ_UNBOUND | WQ_HIGHPRI,
 					 NUM_ACC_QUEUE);
-	if (!gt->usm.acc_wq)
+	if (!gt->usm.acc_wq) {
+		destroy_workqueue(gt->usm.pf_wq);
 		return -ENOMEM;
+	}
 
-	return 0;
+	return devm_add_action_or_reset(xe->drm.dev, pagefault_fini, gt);
 }
 
 void xe_gt_pagefault_reset(struct xe_gt *gt)

drivers/gpu/drm/xe/xe_gt_types.h

@@ -362,6 +362,12 @@ struct xe_gt {
 	 */
 	spinlock_t mcr_lock;
 
+	/**
+	 * @global_invl_lock: protects the register for the duration
+	 *    of a global invalidation of l2 cache
+	 */
+	spinlock_t global_invl_lock;
+
 	/** @wa_active: keep track of active workarounds */
 	struct {
 		/** @wa_active.gt: bitmap with active GT workarounds */

drivers/gpu/drm/xe/xe_guc_submit.c

@@ -284,7 +284,7 @@ static void guc_submit_fini(struct drm_device *drm, void *arg)
 	free_submit_wq(guc);
 }
 
-static void guc_submit_wedged_fini(struct drm_device *drm, void *arg)
+static void guc_submit_wedged_fini(void *arg)
 {
 	struct xe_guc *guc = arg;
 	struct xe_exec_queue *q;
@@ -877,7 +877,7 @@ void xe_guc_submit_wedge(struct xe_guc *guc)
 
 	xe_gt_assert(guc_to_gt(guc), guc_to_xe(guc)->wedged.mode);
 
-	err = drmm_add_action_or_reset(&guc_to_xe(guc)->drm,
+	err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
 				       guc_submit_wedged_fini, guc);
 	if (err) {
 		drm_err(&xe->drm, "Failed to register xe_guc_submit clean-up on wedged.mode=2. Although device is wedged.\n");

drivers/gpu/drm/xe/xe_hw_fence.c

@@ -148,20 +148,20 @@ static const char *xe_hw_fence_get_driver_name(struct dma_fence *dma_fence)
 {
 	struct xe_hw_fence *fence = to_xe_hw_fence(dma_fence);
 
-	return dev_name(gt_to_xe(fence->ctx->gt)->drm.dev);
+	return dev_name(fence->xe->drm.dev);
 }
 
 static const char *xe_hw_fence_get_timeline_name(struct dma_fence *dma_fence)
 {
 	struct xe_hw_fence *fence = to_xe_hw_fence(dma_fence);
 
-	return fence->ctx->name;
+	return fence->name;
 }
 
 static bool xe_hw_fence_signaled(struct dma_fence *dma_fence)
 {
 	struct xe_hw_fence *fence = to_xe_hw_fence(dma_fence);
-	struct xe_device *xe = gt_to_xe(fence->ctx->gt);
+	struct xe_device *xe = fence->xe;
 	u32 seqno = xe_map_rd(xe, &fence->seqno_map, 0, u32);
 
 	return dma_fence->error ||
@@ -253,7 +253,8 @@ void xe_hw_fence_init(struct dma_fence *fence, struct xe_hw_fence_ctx *ctx,
 	struct  xe_hw_fence *hw_fence =
 		container_of(fence, typeof(*hw_fence), dma);
 
-	hw_fence->ctx = ctx;
+	hw_fence->xe = gt_to_xe(ctx->gt);
+	snprintf(hw_fence->name, sizeof(hw_fence->name), "%s", ctx->name);
 	hw_fence->seqno_map = seqno_map;
 	INIT_LIST_HEAD(&hw_fence->irq_link);
 

drivers/gpu/drm/xe/xe_hw_fence_types.h

@@ -12,6 +12,7 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 
+struct xe_device;
 struct xe_gt;
 
 /**
@@ -61,8 +62,10 @@ struct xe_hw_fence_ctx {
 struct xe_hw_fence {
 	/** @dma: base dma fence for hardware fence context */
 	struct dma_fence dma;
-	/** @ctx: hardware fence context */
-	struct xe_hw_fence_ctx *ctx;
+	/** @xe: Xe device for hw fence driver name */
+	struct xe_device *xe;
+	/** @name: name of hardware fence context */
+	char name[MAX_FENCE_NAME_LEN];
 	/** @seqno_map: I/O map for seqno */
 	struct iosys_map seqno_map;
 	/** @irq_link: Link in struct xe_hw_fence_irq.pending */

drivers/gpu/drm/xe/xe_mmio.c

@@ -30,7 +30,8 @@ static void tiles_fini(void *arg)
 	int id;
 
 	for_each_tile(tile, xe, id)
-		tile->mmio.regs = NULL;
+		if (tile != xe_device_get_root_tile(xe))
+			tile->mmio.regs = NULL;
 }
 
 int xe_mmio_probe_tiles(struct xe_device *xe)
@@ -91,9 +92,11 @@ int xe_mmio_probe_tiles(struct xe_device *xe)
 static void mmio_fini(void *arg)
 {
 	struct xe_device *xe = arg;
+	struct xe_tile *root_tile = xe_device_get_root_tile(xe);
 
 	pci_iounmap(to_pci_dev(xe->drm.dev), xe->mmio.regs);
 	xe->mmio.regs = NULL;
+	root_tile->mmio.regs = NULL;
 }
 
 int xe_mmio_init(struct xe_device *xe)
@@ -121,12 +124,29 @@ int xe_mmio_init(struct xe_device *xe)
 	return devm_add_action_or_reset(xe->drm.dev, mmio_fini, xe);
 }
 
+static void mmio_flush_pending_writes(struct xe_gt *gt)
+{
+#define DUMMY_REG_OFFSET	0x130030
+	struct xe_tile *tile = gt_to_tile(gt);
+	int i;
+
+	if (tile->xe->info.platform != XE_LUNARLAKE)
+		return;
+
+	/* 4 dummy writes */
+	for (i = 0; i < 4; i++)
+		writel(0, tile->mmio.regs + DUMMY_REG_OFFSET);
+}
+
 u8 xe_mmio_read8(struct xe_gt *gt, struct xe_reg reg)
 {
 	struct xe_tile *tile = gt_to_tile(gt);
 	u32 addr = xe_mmio_adjusted_addr(gt, reg.addr);
 	u8 val;
 
+	/* Wa_15015404425 */
+	mmio_flush_pending_writes(gt);
+
 	val = readb((reg.ext ? tile->mmio_ext.regs : tile->mmio.regs) + addr);
 	trace_xe_reg_rw(gt, false, addr, val, sizeof(val));
 
@@ -139,6 +159,9 @@ u16 xe_mmio_read16(struct xe_gt *gt, struct xe_reg reg)
 	u32 addr = xe_mmio_adjusted_addr(gt, reg.addr);
 	u16 val;
 
+	/* Wa_15015404425 */
+	mmio_flush_pending_writes(gt);
+
 	val = readw((reg.ext ? tile->mmio_ext.regs : tile->mmio.regs) + addr);
 	trace_xe_reg_rw(gt, false, addr, val, sizeof(val));
 
@@ -160,6 +183,9 @@ u32 xe_mmio_read32(struct xe_gt *gt, struct xe_reg reg)
 	u32 addr = xe_mmio_adjusted_addr(gt, reg.addr);
 	u32 val;
 
+	/* Wa_15015404425 */
+	mmio_flush_pending_writes(gt);
+
 	if (!reg.vf && IS_SRIOV_VF(gt_to_xe(gt)))
 		val = xe_gt_sriov_vf_read32(gt, reg);
 	else

drivers/gpu/drm/xe/xe_observation.c

@@ -66,7 +66,6 @@ static struct ctl_table observation_ctl_table[] = {
 	 .extra1 = SYSCTL_ZERO,
 	 .extra2 = SYSCTL_ONE,
 	 },
-	{}
 };
 
 /**

drivers/gpu/drm/xe/xe_pat.c

@@ -7,6 +7,8 @@
 
 #include <drm/xe_drm.h>
 
+#include <generated/xe_wa_oob.h>
+
 #include "regs/xe_reg_defs.h"
 #include "xe_assert.h"
 #include "xe_device.h"
@@ -15,6 +17,7 @@
 #include "xe_gt_mcr.h"
 #include "xe_mmio.h"
 #include "xe_sriov.h"
+#include "xe_wa.h"
 
 #define _PAT_ATS				0x47fc
 #define _PAT_INDEX(index)			_PICK_EVEN_2RANGES(index, 8, \
@@ -382,7 +385,13 @@ void xe_pat_init_early(struct xe_device *xe)
 	if (GRAPHICS_VER(xe) == 20) {
 		xe->pat.ops = &xe2_pat_ops;
 		xe->pat.table = xe2_pat_table;
-		xe->pat.n_entries = ARRAY_SIZE(xe2_pat_table);
+
+		/* Wa_16023588340. XXX: Should use XE_WA */
+		if (GRAPHICS_VERx100(xe) == 2001)
+			xe->pat.n_entries = 28; /* Disable CLOS3 */
+		else
+			xe->pat.n_entries = ARRAY_SIZE(xe2_pat_table);
+
 		xe->pat.idx[XE_CACHE_NONE] = 3;
 		xe->pat.idx[XE_CACHE_WT] = 15;
 		xe->pat.idx[XE_CACHE_WB] = 2;

drivers/gpu/drm/xe/xe_pm.c

@@ -91,13 +91,13 @@ int xe_pm_suspend(struct xe_device *xe)
 	for_each_gt(gt, xe, id)
 		xe_gt_suspend_prepare(gt);
 
+	xe_display_pm_suspend(xe, false);
+
 	/* FIXME: Super racey... */
 	err = xe_bo_evict_all(xe);
 	if (err)
 		goto err;
 
-	xe_display_pm_suspend(xe, false);
-
 	for_each_gt(gt, xe, id) {
 		err = xe_gt_suspend(gt);
 		if (err) {
@@ -151,11 +151,11 @@ int xe_pm_resume(struct xe_device *xe)
 
 	xe_irq_resume(xe);
 
-	xe_display_pm_resume(xe, false);
-
 	for_each_gt(gt, xe, id)
 		xe_gt_resume(gt);
 
+	xe_display_pm_resume(xe, false);
+
 	err = xe_bo_restore_user(xe);
 	if (err)
 		goto err;
@@ -363,10 +363,11 @@ int xe_pm_runtime_suspend(struct xe_device *xe)
 	mutex_unlock(&xe->mem_access.vram_userfault.lock);
 
 	if (xe->d3cold.allowed) {
+		xe_display_pm_suspend(xe, true);
+
 		err = xe_bo_evict_all(xe);
 		if (err)
 			goto out;
-		xe_display_pm_suspend(xe, true);
 	}
 
 	for_each_gt(gt, xe, id) {

drivers/gpu/drm/xe/xe_preempt_fence.c

@@ -128,8 +128,9 @@ xe_preempt_fence_arm(struct xe_preempt_fence *pfence, struct xe_exec_queue *q,
 {
 	list_del_init(&pfence->link);
 	pfence->q = xe_exec_queue_get(q);
+	spin_lock_init(&pfence->lock);
 	dma_fence_init(&pfence->base, &preempt_fence_ops,
-		      &q->lr.lock, context, seqno);
+		      &pfence->lock, context, seqno);
 
 	return &pfence->base;
 }

drivers/gpu/drm/xe/xe_preempt_fence_types.h

@@ -25,6 +25,8 @@ struct xe_preempt_fence {
 	struct xe_exec_queue *q;
 	/** @preempt_work: work struct which issues preemption */
 	struct work_struct preempt_work;
+	/** @lock: dma-fence fence lock */
+	spinlock_t lock;
 	/** @error: preempt fence is in error state */
 	int error;
 };

drivers/gpu/drm/xe/xe_sched_job.c

@@ -171,12 +171,13 @@ void xe_sched_job_destroy(struct kref *ref)
 	struct xe_sched_job *job =
 		container_of(ref, struct xe_sched_job, refcount);
 	struct xe_device *xe = job_to_xe(job);
+	struct xe_exec_queue *q = job->q;
 
 	xe_sched_job_free_fences(job);
-	xe_exec_queue_put(job->q);
 	dma_fence_put(job->fence);
 	drm_sched_job_cleanup(&job->drm);
 	job_free(job);
+	xe_exec_queue_put(q);
 	xe_pm_runtime_put(xe);
 }
 

drivers/gpu/drm/xe/xe_trace.h

@@ -309,7 +309,7 @@ DECLARE_EVENT_CLASS(xe_hw_fence,
 		    TP_ARGS(fence),
 
 		    TP_STRUCT__entry(
-			     __string(dev, __dev_name_gt(fence->ctx->gt))
+			     __string(dev, __dev_name_xe(fence->xe))
 			     __field(u64, ctx)
 			     __field(u32, seqno)
 			     __field(struct xe_hw_fence *, fence)

drivers/gpu/drm/xe/xe_wa.c

@@ -486,6 +486,10 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	  XE_RTP_RULES(GRAPHICS_VERSION(2004), FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, SLM_WMTP_RESTORE))
 	},
+	{ XE_RTP_NAME("14021402888"),
+	  XE_RTP_RULES(GRAPHICS_VERSION(2004), ENGINE_CLASS(RENDER)),
+	  XE_RTP_ACTIONS(SET(HALF_SLICE_CHICKEN7, CLEAR_OPTIMIZATION_DISABLE))
+	},
 
 	/* Xe2_HPG */
 
@@ -538,6 +542,20 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	  XE_RTP_RULES(GRAPHICS_VERSION(2001), ENGINE_CLASS(RENDER)),
 	  XE_RTP_ACTIONS(SET(HALF_SLICE_CHICKEN7, CLEAR_OPTIMIZATION_DISABLE))
 	},
+	{ XE_RTP_NAME("14021821874"),
+	  XE_RTP_RULES(GRAPHICS_VERSION(2001), FUNC(xe_rtp_match_first_render_or_compute)),
+	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, STK_ID_RESTRICT))
+	},
+
+	/* Xe2_LPM */
+
+	{ XE_RTP_NAME("16021639441"),
+	  XE_RTP_RULES(MEDIA_VERSION(2000)),
+	  XE_RTP_ACTIONS(SET(CSFE_CHICKEN1(0),
+			     GHWSP_CSB_REPORT_DIS |
+			     PPHWSP_CSB_AND_TIMESTAMP_REPORT_DIS,
+			     XE_RTP_ACTION_FLAG(ENGINE_BASE)))
+	},
 
 	/* Xe2_HPM */
 

drivers/gpu/drm/xe/xe_wa_oob.rules

@@ -29,3 +29,4 @@
 13011645652	GRAPHICS_VERSION(2004)
 22019338487	MEDIA_VERSION(2000)
 		GRAPHICS_VERSION(2001)
+16023588340	GRAPHICS_VERSION(2001)

include/uapi/drm/xe_drm.h

@@ -1590,10 +1590,10 @@ enum drm_xe_oa_property_id {
 	 * b. Counter select c. Counter size and d. BC report. Also refer to the
 	 * oa_formats array in drivers/gpu/drm/xe/xe_oa.c.
 	 */
-#define DRM_XE_OA_FORMAT_MASK_FMT_TYPE		(0xff << 0)
-#define DRM_XE_OA_FORMAT_MASK_COUNTER_SEL	(0xff << 8)
-#define DRM_XE_OA_FORMAT_MASK_COUNTER_SIZE	(0xff << 16)
-#define DRM_XE_OA_FORMAT_MASK_BC_REPORT		(0xff << 24)
+#define DRM_XE_OA_FORMAT_MASK_FMT_TYPE		(0xffu << 0)
+#define DRM_XE_OA_FORMAT_MASK_COUNTER_SEL	(0xffu << 8)
+#define DRM_XE_OA_FORMAT_MASK_COUNTER_SIZE	(0xffu << 16)
+#define DRM_XE_OA_FORMAT_MASK_BC_REPORT		(0xffu << 24)
 
 	/**
 	 * @DRM_XE_OA_PROPERTY_OA_PERIOD_EXPONENT: Requests periodic OA unit