Claude review: drm/ttm: Use common ancestor of evictor and evictee as limit pool

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

Patch Review

This patch fixes protection calculation for sibling cgroups by using their common ancestor as the limit pool root. Without this, recursive protection propagation makes all children of a protected parent appear protected, preventing a protected sibling from evicting an unprotected sibling.

**The self-eviction guard**:
```c
if (!evict_walk->alloc_state->may_try_low &&
        bo->resource->css == evict_walk->alloc_state->charge_pool)
    return 0;
```
This prevents a protected cgroup from evicting its own buffers during the "aggressive protected" eviction path. The pointer comparison works because `charge_pool` was obtained from `get_current_dmemcs()` and `bo->resource->css` is the pool charged during allocation. However, this is a **pointer comparison** — if two different pool_state objects refer to the same cgroup (e.g., obtained at different times), this could fail to match. Looking at `dmem_cgroup_try_charge`, the pool is obtained via `get_cg_pool_unlocked` which caches pool states per cgroup, so the same cgroup should yield the same pointer. This should be safe.

**The ancestor lookup**:
```c
if (!limit_pool) {
    ancestor = dmem_cgroup_get_common_ancestor(bo->resource->css,
                           evict_walk->alloc_state->charge_pool);
    limit_pool = ancestor;
}
```
When there's no explicit limit_pool (the aggressive protection path), the common ancestor is computed per-eviction-candidate. This is O(depth) per candidate, which seems acceptable for typical cgroup hierarchies. The reference is properly put after the `dmem_cgroup_state_evict_valuable` call.

**Indentation**: The `&&` continuation at line 1852 uses tabs for alignment rather than aligning under the `(`. This may or may not match local style; the existing code in this file seems to use tab-based continuation which is kernel style.

Overall this is a well-thought-out series. The main actionable items are:
1. The `@b: First pool` → `@b: Second pool` typo in patch 2
2. The potential RCU concern around `cgroup_e_css()` usage in `dmem_cgroup_get_common_ancestor()` in patch 2
3. The "pendant" → "counterpart" wording in patch 1

---
Generated by Claude Code Patch Reviewer