Claude review: drm/hisilicon/hibmc: use clock to look up the PLL value

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

Patch Review

**1. Unused `CLOCK_TOLERANCE` define:** The patch adds `#define CLOCK_TOLERANCE 100` but never uses it. The tolerance check in `hibmc_get_best_clock_idx()` uses `mode->clock / 100` (1% tolerance) instead. The `CLOCK_TOLERANCE` define should either be removed or actually used.

**2. Tolerance logic for low clocks:** The expression `diff < mode->clock / 100` uses integer division. For the 25000 kHz entry, the tolerance is 250 kHz which is fine. But if a mode somehow had a very low clock, the tolerance could round to 0 due to integer division, making the match impossible. Not a practical problem given the table entries, but `<= 0` should be guarded (e.g., `mode->clock / 100` should be at least 1). 

**3. `get_pll_config` now takes non-const mode pointer:**
```c
-static void get_pll_config(u64 x, u64 y, u32 *pll1, u32 *pll2)
+static void get_pll_config(struct drm_display_mode *mode, u32 *pll1, u32 *pll2)
```
The `mode` parameter should be `const struct drm_display_mode *mode` since it's not modified. The `hibmc_get_best_clock_idx` already takes a `const` pointer, so this is just a missed annotation on `get_pll_config`.

**4. Good refactoring:** Consolidating the table lookup into `hibmc_get_best_clock_idx()` and reusing it in both `hibmc_crtc_mode_valid()` and `get_pll_config()` is a nice cleanup that eliminates code duplication.

**5. `hibmc_dislay_pll_config` typo:** The struct name `hibmc_dislay_pll_config` has a typo — it should be `hibmc_display_pll_config`. This is pre-existing and not introduced by this patch, but could be cleaned up while touching this code.

---
Generated by Claude Code Patch Reviewer