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 3CD4D3CD8BA for ; Mon, 16 Mar 2026 16:45:04 +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=1773679504; cv=none; b=dPDiBqaTvS+kKM7s3qkoYpvlzGzLn6UlAlHg1iCjXvkPzPX+eqsxDYw2GazvegeEIUhVTRd0zxbt2JhcD18r0axVn5jHGPk2ACniLdshrSJjnv+T29grXofT20zkk4HX+2ejQCoi1MlAW4ngas8zl8LSj0RCyAsvWtY0rlNTcRQ= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773679504; c=relaxed/simple; bh=jZgP9vRbn/1PtWSc68dmfAY0rH/kP7mm4Lhq1dZwZrA=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=L1rQrGkZou5TtcEt8u+kY8oBxCSG4QkdqLkTPd+g78yc7k5LOdXgU+THZscbMyXSdjb2zvYdiOF1Ow8sQHFnx1xphagqhXwDEmTEALCpNWkLvbhTFwKhQurhQCUCcm4dHwLAcsEOQdJwn2bnJdydKDBbkPqk8cXkOxriDE3irJc= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=OUKO1zSx; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="OUKO1zSx" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 6974DC2BC87; Mon, 16 Mar 2026 16:45:03 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1773679503; bh=jZgP9vRbn/1PtWSc68dmfAY0rH/kP7mm4Lhq1dZwZrA=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=OUKO1zSxhSPjWmw3bANKzs2NOLX+6QSJJ2BfFDhhvDsW4uX5oQeqJ1klVEFm/N658 BVRltk5AEbQqTP2C8XT4JoDNnqf8EE3fSGDVEcXzwq6QAJKxJykYWHWHCRK2EZXwpY wKhxEQYAawqf6OhwKHCa4A8qTTL4O+67aGAABDaT2zT94HbOrk8uDgtEDDQWnEnNys dvSNMWH+CC3Oud7EMBooCN4F5+Ee+kwS1IrDZmZUpt7Ru/YLe4gM4mt/BIMqpYl4P5 lAsYbVFfgbsjyx2uhqn6VDHX7peUAAl4SgNaEpTPDy4wLuqc6kp7Ce+d1djoq3IqL7 GbTVdVjvg5DUA== From: Sasha Levin To: stable@vger.kernel.org Cc: Mehul Rao , Takashi Iwai , Sasha Levin Subject: [PATCH 5.15.y 2/2] ALSA: pcm: fix use-after-free on linked stream runtime in snd_pcm_drain() Date: Mon, 16 Mar 2026 12:44:59 -0400 Message-ID: <20260316164459.944343-2-sashal@kernel.org> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20260316164459.944343-1-sashal@kernel.org> References: <2026031626-glowworm-observant-79df@gregkh> <20260316164459.944343-1-sashal@kernel.org> 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 From: Mehul Rao [ Upstream commit 9b1dbd69ba6f8f8c69bc7b77c2ce3b9c6ed05ba6 ] 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: Sasha Levin --- sound/core/pcm_native.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index f55f476b0a148..1680f542e79ab 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -2144,6 +2144,10 @@ static int snd_pcm_drain(struct snd_pcm_substream *substream, 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_substream *substream, 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); -- 2.51.0