Claude review: drm: Suppress intentional warning backtraces in scaling unit tests

[Claude Code Review Bot <claude-review@example.com>]

Patch Review

**Overall: Good practical demonstration.**

**Wrapping non-warning cases in suppression is harmless but unnecessary:**

```c
+	kunit_warning_suppress(test) {
+		scaling_factor = drm_rect_calc_hscale(&params->src, &params->dst,
+						      params->min_range,
+						      params->max_range);
+		KUNIT_EXPECT_SUPPRESSED_WARNING_COUNT(test, expected_warnings);
+	}
```

When `expected_warnings == 0` (the non-error cases), the entire call is wrapped in a suppression block unnecessarily. This adds trivial overhead (kzalloc, list add/remove) for the non-warning parameterized cases. An alternative would be to conditionally wrap only the `-EINVAL` cases, but the current approach is simpler and more readable, and the overhead is negligible for tests. This is a reasonable trade-off.

**`INT_MIN` initialization:**

```c
+	int scaling_factor = INT_MIN;
```

The v10 changelog explains this was added to avoid "potentially uninitialized" warnings. This works but is defensive coding since `scaling_factor` is always assigned inside the block before the `KUNIT_EXPECT_EQ`. Still, it's harmless and the compiler might not see through the `for`-loop-based macro.

**`#include <linux/limits.h>` for `INT_MIN`:** Correct — `INT_MIN` is defined there.

**CONFIG_BUG skip is correct:**

```c
+	if (expected_warnings && !IS_ENABLED(CONFIG_BUG))
+		kunit_skip(test, "requires CONFIG_BUG");
```

Without `CONFIG_BUG`, WARN_ON becomes a no-op and the count would be 0, failing the count expectation. Skipping only the warning-expected parameterized cases is the right behavior.

---
Generated by Claude Code Patch Reviewer