[PATCH v5 0/5] mm: Unconditional per-VMA locks and cleanups · Suren Baghdasaryan <surenb@google.com> · 2026-08-12
[PATCH v5 1/5] mm: Make per-VMA locks available universally · Suren Baghdasaryan <surenb@google.com> · 2026-08-12
[PATCH v5 2/5] binder: Make shrinker rely solely on per-VMA lock · Suren Baghdasaryan <surenb@google.com> · 2026-08-12
Re: [PATCH v5 2/5] binder: Make shrinker rely solely on per-VMA lock · Alice Ryhl <aliceryhl@google.com> · 2026-08-12
Re: [PATCH v5 2/5] binder: Make shrinker rely solely on per-VMA lock · Suren Baghdasaryan <surenb@google.com> · 2026-08-12
[PATCH v5 3/5] mm: Add RCU-based VMA lookup helper that waits for writers · Suren Baghdasaryan <surenb@google.com> · 2026-08-12
[PATCH v5 4/5] binder: Remove mmap_lock fallback · Suren Baghdasaryan <surenb@google.com> · 2026-08-12
[PATCH v5 5/5] tcp: Remove mmap_lock fallback path · Suren Baghdasaryan <surenb@google.com> · 2026-08-12

Re: [PATCH v5 2/5] binder: Make shrinker rely solely on per-VMA lock

From: Alice Ryhl <aliceryhl@google.com>
Date: 2026-08-12 10:46:35
Also in: linux-mm, lkml

On Tue, Aug 11, 2026 at 06:10:52PM -0700, Suren Baghdasaryan wrote:

 	return LRU_REMOVED_RETRY;
 
-err_invalid_vma:
 	mutex_unlock(&alloc->mutex);

With the err_invalid_vma label removed, this call to mutex_unlock() is
no longer reachable as it comes right after a 'return' statement.

Alice