On 10/9/2023 13:02, Andi Shyti wrote:
Hi John,
...
if (intf_id >= INTEL_GSC_NUM_INTERFACES) {
- drm_warn_once(>->i915->drm, "GSC irq: intf_id %d is out of range", intf_id);
+ gt_warn_once(gt, "GSC irq: intf_id %d is out of range", intf_id);
return;
}
if (!HAS_HECI_GSC(gt->i915)) {
- drm_warn_once(>->i915->drm, "GSC irq: not supported");
+ gt_warn_once(gt, "GSC irq: not supported");
return;
}
@@ -300,7 +301,7 @@ static void gsc_irq_handler(struct intel_gt *gt, unsigned int intf_id)
ret = generic_handle_irq(gt->gsc.intf[intf_id].irq);
if (ret)
- drm_err_ratelimited(>->i915->drm, "error handling GSC irq: %d\n", ret);
+ gt_err_ratelimited(gt, "error handling GSC irq: %d\n", ret);
}
void intel_gsc_irq_handler(struct intel_gt *gt, u32 iir)
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_print.h b/drivers/gpu/drm/i915/gt/intel_gt_print.h
index 55a336a9ff061..7fdc78c79273e 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_print.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_print.h
@@ -16,6 +16,9 @@
#define gt_warn(_gt, _fmt, ...) \
drm_warn(&(_gt)->i915->drm, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
+#define gt_warn_once(_gt, _fmt, ...) \
+ drm_warn_once(&(_gt)->i915->drm, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
+
I would add the gt_warn_once() part in a different patch.
But this is the patch that uses it. You should not add dead code. The only
exception being if it is something large and complex that needs to be added
in stages for ease of code review. But this really doesn't count as large or
complex!
I wouldn't call it dead code if it's used right after... you
could also put all the *_warn_* changes in different patch.
I did start splitting it into errors vs debugs but then decided it
wasn't worth the effort ;)
Anyway, I don't have a strong opinion for such a straight forward
patch, so that I'm fine with it as it is:
Reviewed-by: Andi Shyti <andi.shyti@xxxxxxxxxxxxxxx>
Thanks :).
Andi
