[PATCH v1 3/3] drm/xe/cri: Wedge the device on hardware error

[Raag Jadav <raag.jadav@intel.com>]

Hardware error interrupts are routed to SGUnit only for debug cases on
Crescent Island. Wedge the device and have it available to the user for
debugging.

Signed-off-by: Raag Jadav <raag.jadav@intel.com>
---
 drivers/gpu/drm/xe/xe_hw_error.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c
index b704eab7d3bd..b756cbe321d6 100644
--- a/drivers/gpu/drm/xe/xe_hw_error.c
+++ b/drivers/gpu/drm/xe/xe_hw_error.c
@@ -55,11 +55,11 @@ static void hw_error_work(struct work_struct *work)
 {
 	struct xe_tile *tile = container_of(work, typeof(*tile), hw_error_work);
 	struct xe_device *xe = tile_to_xe(tile);
-	int ret;
 
-	ret = xe_survivability_mode_runtime_enable(xe);
-	if (ret)
-		drm_err(&xe->drm, "Failed to enable runtime survivability mode\n");
+	if (xe->info.platform == XE_CRESCENTISLAND)
+		xe_device_declare_wedged(xe);
+	else if (xe->info.platform == XE_BATTLEMAGE)
+		xe_survivability_mode_runtime_enable(xe);
 }
 
 static void csc_hw_error_handler(struct xe_tile *tile, const enum hardware_error hw_err)
@@ -104,6 +104,17 @@ static void hw_error_source_handler(struct xe_tile *tile, const enum hardware_er
 	unsigned long flags;
 	u32 err_src;
 
+	/*
+	 * We should never be here for CRI other than debug cases, so retain the hardware
+	 * state and bail without clearing error registers.
+	 */
+	if (xe->info.platform == XE_CRESCENTISLAND) {
+		drm_err_ratelimited(&xe->drm, HW_ERR "Tile%d reported %s error\n",
+				    tile->id, hw_err_str);
+		schedule_work(&tile->hw_error_work);
+		return;
+	}
+
 	if (xe->info.platform != XE_BATTLEMAGE)
 		return;
 
-- 
2.43.0