Claude review: drm/radeon/radeon_connectors: use struct drm_edid instead of struct edid

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

Patch Review

**Positive aspects:**
- The field type change in `radeon_mode.h` from `struct edid *edid` to `const struct drm_edid *edid` is correct and matches the return type of `drm_edid_read_ddc()` / `drm_edid_read_switcheroo()`.
- The `drm_edid_connector_update()` + `drm_edid_connector_add_modes()` replacement is the correct modern pattern.
- The `drm_edid_is_digital()` replacement for `!!(radeon_connector->edid->input & DRM_EDID_INPUT_DIGITAL)` is correct — `drm_edid_is_digital()` takes `const struct drm_edid *`.
- The `radeon_bios_get_hardcoded_edid()` change from `drm_edid_duplicate(drm_edid_raw(...))` to `drm_edid_dup(...)` is a clean simplification.

**Issue 1 — `radeon_connector_free_edid` left using `kfree` on `struct drm_edid`:**

After patch 1, the field type changes to `const struct drm_edid *`, but `radeon_connector_free_edid()` still calls `kfree()` on it:

```c
static void radeon_connector_free_edid(struct drm_connector *connector)
{
	struct radeon_connector *radeon_connector = to_radeon_connector(connector);

	kfree(radeon_connector->edid);
	radeon_connector->edid = NULL;
}
```

After patch 1, `radeon_connector->edid` is now a `const struct drm_edid *`, but `kfree()` is still used instead of `drm_edid_free()`. This is a **correctness bug between patches** — `struct drm_edid` is allocated by the `drm_edid_read_*` functions and contains an internal `const struct edid *edid` pointer that also needs freeing. Simply calling `kfree()` on the outer `struct drm_edid` will leak the inner EDID data. The tree must compile and be correct at every patch boundary, so patch 1 should update `radeon_connector_free_edid` to call `drm_edid_free()` or the two patches should be squashed.

Additionally, `kfree()` on a `const` pointer will generate a compiler warning/error about discarding the `const` qualifier.

**Issue 2 — `drm_edid_to_sad` / `drm_edid_to_speaker_allocation` take `const struct edid *`:**

The audio code changes are correct. These functions take `const struct edid *`, so using `drm_edid_raw()` to extract the raw EDID is the right approach:

```c
sad_count = drm_edid_to_sad(drm_edid_raw(radeon_connector->edid), &sads);
sad_count = drm_edid_to_speaker_allocation(drm_edid_raw(radeon_connector->edid), &sadb);
```

No issue here, just confirming it's correct.

**Issue 3 — `drm_edid_connector_update` return value ignored:**

In `radeon_ddc_get_modes`, the return value of `drm_edid_connector_update()` is silently discarded:

```c
	if (radeon_connector->edid) {
		drm_edid_connector_update(connector, radeon_connector->edid);
		ret = drm_edid_connector_add_modes(connector);
		return ret;
	}
	drm_edid_connector_update(connector, NULL);
```

The old `drm_connector_update_edid_property()` also returned int and was also ignored, so this is pre-existing behavior and not a regression. Just noting it.

**Nit — `drm_edid_read_ddc` vs `drm_edid_read`:**

For the non-DP, non-switcheroo, plain i2c adapter case, `drm_edid_read()` could be used instead of `drm_edid_read_ddc()` since `drm_edid_read()` is just `drm_edid_read_ddc(connector, connector->ddc)`. However, radeon sets the ddc_bus separately from the connector's `ddc` field, so using `drm_edid_read_ddc` with the explicit adapter is correct here.

---

---
Generated by Claude Code Patch Reviewer