From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4C5E112B94; Tue, 17 Mar 2026 16:44:52 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773765892; cv=none; b=AFF3b/jZRirKJlxmibVb1u3R9FNCQJRy7+hmnNb3BYQEvjpc3c4QSnDwAXNpxn+mLjpqTsJefMkO4At7Re3E0Hxxaon81G0MCYByr20kJWAohAQsS0a6Fy8xy2qZsuC4rh2Od06DZ9i4t/xTvKHAjgiX61b1RHXdUNc0ytuc7Q0= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773765892; c=relaxed/simple; bh=3R9G1ynX+FdK43fFQIwf3aXuS4yYp6yB18su9yNKiCs=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=RXfD2jrtmsvbFypfrCj4GR1wL/Yyjj7ZTnv4SAA8CTo4ZTQNb+aTMFe2ziGxtWNaAfVaGUdYoRI7ywZD4QC6i7JaNHY4aFi+sdEHtd8BIzgJU9yx1nRN5c9KefzRIZLmLDJdJwuiSMT8BvPCcwTx5ZsdvCjZWYZbGIyZnGmR6aw= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=lJxzbp7v; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="lJxzbp7v" Received: by smtp.kernel.org (Postfix) with ESMTPSA id B48E4C4CEF7; Tue, 17 Mar 2026 16:44:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1773765892; bh=3R9G1ynX+FdK43fFQIwf3aXuS4yYp6yB18su9yNKiCs=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=lJxzbp7vmVipjrCKE2/WMv5SkOh2dcuTM8aAkEPYMdeG1Wg3It+x1vq/7S7yjg84S ZCLaTZ7byfE8jnFDJm8WbbsOXrtfexKLSB27eMsNoH9oJzYBU6hSlfxR/qyruQVX91 UfXdwBwtTEWS3j3n9R/FjEaLoveql/Qc6GFeiwIw= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Mehul Rao , Takashi Iwai Subject: [PATCH 6.19 118/378] ALSA: pcm: fix use-after-free on linked stream runtime in snd_pcm_drain() Date: Tue, 17 Mar 2026 17:31:15 +0100 Message-ID: <20260317163011.360102000@linuxfoundation.org> X-Mailer: git-send-email 2.53.0 In-Reply-To: <20260317163006.959177102@linuxfoundation.org> References: <20260317163006.959177102@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: stable@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 6.19-stable review patch. If anyone has any objections, please let me know. ------------------ From: Mehul Rao commit 9b1dbd69ba6f8f8c69bc7b77c2ce3b9c6ed05ba6 upstream. In the drain loop, the local variable 'runtime' is reassigned to a linked stream's runtime (runtime = s->runtime at line 2157). After releasing the stream lock at line 2169, the code accesses runtime->no_period_wakeup, runtime->rate, and runtime->buffer_size (lines 2170-2178) — all referencing the linked stream's runtime without any lock or refcount protecting its lifetime. A concurrent close() on the linked stream's fd triggers snd_pcm_release_substream() → snd_pcm_drop() → pcm_release_private() → snd_pcm_unlink() → snd_pcm_detach_substream() → kfree(runtime). No synchronization prevents kfree(runtime) from completing while the drain path dereferences the stale pointer. Fix by caching the needed runtime fields (no_period_wakeup, rate, buffer_size) into local variables while still holding the stream lock, and using the cached values after the lock is released. Fixes: f2b3614cefb6 ("ALSA: PCM - Don't check DMA time-out too shortly") Cc: stable@vger.kernel.org Signed-off-by: Mehul Rao Link: https://patch.msgid.link/20260305193508.311096-1-mehulrao@gmail.com Signed-off-by: Takashi Iwai Signed-off-by: Greg Kroah-Hartman --- sound/core/pcm_native.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -2144,6 +2144,10 @@ static int snd_pcm_drain(struct snd_pcm_ for (;;) { long tout; struct snd_pcm_runtime *to_check; + unsigned int drain_rate; + snd_pcm_uframes_t drain_bufsz; + bool drain_no_period_wakeup; + if (signal_pending(current)) { result = -ERESTARTSYS; break; @@ -2163,16 +2167,25 @@ static int snd_pcm_drain(struct snd_pcm_ snd_pcm_group_unref(group, substream); if (!to_check) break; /* all drained */ + /* + * Cache the runtime fields needed after unlock. + * A concurrent close() on the linked stream may free + * its runtime via snd_pcm_detach_substream() once we + * release the stream lock below. + */ + drain_no_period_wakeup = to_check->no_period_wakeup; + drain_rate = to_check->rate; + drain_bufsz = to_check->buffer_size; init_waitqueue_entry(&wait, current); set_current_state(TASK_INTERRUPTIBLE); add_wait_queue(&to_check->sleep, &wait); snd_pcm_stream_unlock_irq(substream); - if (runtime->no_period_wakeup) + if (drain_no_period_wakeup) tout = MAX_SCHEDULE_TIMEOUT; else { tout = 100; - if (runtime->rate) { - long t = runtime->buffer_size * 1100 / runtime->rate; + if (drain_rate) { + long t = drain_bufsz * 1100 / drain_rate; tout = max(t, tout); } tout = msecs_to_jiffies(tout);