drm/xe/wa: Track gt/engine/lrc active workarounds

Allocate the data to track workarounds on each gt of the device,
and pass that to RTP so the active workarounds are tracked.

Even if the workarounds available until now are mostly device
or platform centric, with the different IP versions for media and
graphics starting with MTL, it's possible that some workarounds
need to be applied only on select GTs. Also, given the workaround
database is per IP block, for tracking purposes there is no need to
differentiate the workarounds per engine class. Hence the bitmask
to track active workarounds can be tracked per GT.

v2: Move the tracking from per-device to per-GT basis (Matt Roper)

Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20230526164358.86393-7-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

drivers/gpu/drm/xe/xe_gt.c

@@ -315,6 +315,11 @@ int xe_gt_init_early(struct xe_gt *gt)
 		return err;
 
 	xe_reg_sr_init(&gt->reg_sr, "GT", gt_to_xe(gt));
+
+	err = xe_wa_init(gt);
+	if (err)
+		return err;
+
 	xe_wa_process_gt(gt);
 	xe_tuning_process_gt(gt);
 

drivers/gpu/drm/xe/xe_gt_types.h

@@ -359,6 +359,16 @@ struct xe_gt {
 	 *    of a steered operation
 	 */
 	spinlock_t mcr_lock;
+
+	/** @wa_active: keep track of active workarounds */
+	struct {
+		/** @gt: bitmap with active GT workarounds */
+		unsigned long *gt;
+		/** @engine: bitmap with active engine workarounds */
+		unsigned long *engine;
+		/** @lrc: bitmap with active LRC workarounds */
+		unsigned long *lrc;
+	} wa_active;
 };
 
 #endif

drivers/gpu/drm/xe/xe_wa.c

@@ -5,6 +5,7 @@
 
 #include "xe_wa.h"
 
+#include <drm/drm_managed.h>
 #include <kunit/visibility.h>
 #include <linux/compiler_types.h>
 
@@ -581,6 +582,8 @@ void xe_wa_process_gt(struct xe_gt *gt)
 {
 	struct xe_rtp_process_ctx ctx = XE_RTP_PROCESS_CTX_INITIALIZER(gt);
 
+	xe_rtp_process_ctx_enable_active_tracking(&ctx, gt->wa_active.gt,
+						  ARRAY_SIZE(gt_was));
 	xe_rtp_process_to_sr(&ctx, gt_was, &gt->reg_sr);
 }
 EXPORT_SYMBOL_IF_KUNIT(xe_wa_process_gt);
@@ -597,6 +600,8 @@ void xe_wa_process_engine(struct xe_hw_engine *hwe)
 {
 	struct xe_rtp_process_ctx ctx = XE_RTP_PROCESS_CTX_INITIALIZER(hwe);
 
+	xe_rtp_process_ctx_enable_active_tracking(&ctx, hwe->gt->wa_active.engine,
+						  ARRAY_SIZE(engine_was));
 	xe_rtp_process_to_sr(&ctx, engine_was, &hwe->reg_sr);
 }
 
@@ -612,5 +617,37 @@ void xe_wa_process_lrc(struct xe_hw_engine *hwe)
 {
 	struct xe_rtp_process_ctx ctx = XE_RTP_PROCESS_CTX_INITIALIZER(hwe);
 
+	xe_rtp_process_ctx_enable_active_tracking(&ctx, hwe->gt->wa_active.lrc,
+						  ARRAY_SIZE(lrc_was));
 	xe_rtp_process_to_sr(&ctx, lrc_was, &hwe->reg_lrc);
 }
+
+/**
+ * xe_wa_init - initialize gt with workaround bookkeeping
+ * @gt: GT instance to initialize
+ *
+ * Returns 0 for success, negative error code otherwise.
+ */
+int xe_wa_init(struct xe_gt *gt)
+{
+	struct xe_device *xe = gt_to_xe(gt);
+	size_t n_lrc, n_engine, n_gt, total;
+	unsigned long *p;
+
+	n_gt = BITS_TO_LONGS(ARRAY_SIZE(gt_was));
+	n_engine = BITS_TO_LONGS(ARRAY_SIZE(engine_was));
+	n_lrc = BITS_TO_LONGS(ARRAY_SIZE(lrc_was));
+	total = n_gt + n_engine + n_lrc;
+
+	p = drmm_kzalloc(&xe->drm, sizeof(*p) * total, GFP_KERNEL);
+	if (!p)
+		return -ENOMEM;
+
+	gt->wa_active.gt = p;
+	p += n_gt;
+	gt->wa_active.engine = p;
+	p += n_engine;
+	gt->wa_active.lrc = p;
+
+	return 0;
+}

drivers/gpu/drm/xe/xe_wa.h

@@ -9,6 +9,7 @@
 struct xe_gt;
 struct xe_hw_engine;
 
+int xe_wa_init(struct xe_gt *gt);
 void xe_wa_process_gt(struct xe_gt *gt);
 void xe_wa_process_engine(struct xe_hw_engine *hwe);
 void xe_wa_process_lrc(struct xe_hw_engine *hwe);