From: Sasha Levin <sashal@kernel.org>
To: Takashi Iwai <tiwai@suse.com>
Cc: Sasha Levin <sashal@kernel.org>, Jaroslav Kysela <perex@perex.cz>,
Zhao Dongdong <zhaodongdong@kylinos.cn>,
linux-sound@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH] ALSA: hda/core: Fix double free on stream page allocation failure
Date: Wed, 28 Oct 2026 10:44:43 -0400 [thread overview]
Message-ID: <20261028144443.992569-1-sashal@kernel.org> (raw)
Since commit f16eaa38ea64 ("ALSA: hda/core: add cleanup in
snd_hdac_bus_alloc_stream_pages()"), snd_hdac_bus_alloc_stream_pages()
frees the BDL and position buffers it has already allocated when a
later allocation fails. snd_dma_free_pages() doesn't clear dmab->area,
so the freed buffers still look allocated to
snd_hdac_bus_free_stream_pages(), which the HDA controller drivers call
from their destructors, and they are freed a second time when the
device is removed.
With fault injection failing the CORB/RIRB buffer allocation of
snd-hda-intel on a QEMU ich9-intel-hda, unbinding the device then
crashes:
page dumped because: VM_BUG_ON_PAGE(page_ref_count(page) == 0)
kernel BUG at include/linux/mm.h:1986!
Call Trace:
___free_pages+0x9b/0xa0
free_pages_exact+0xa5/0xd0
snd_hdac_bus_free_stream_pages+0x43/0xb0 [snd_hda_core]
azx_free+0xb3/0x1b0 [snd_hda_intel]
azx_dev_free+0xd/0x20 [snd_hda_intel]
Clear the buffer pointers after freeing them in the error path.
Fixes: f16eaa38ea64 ("ALSA: hda/core: add cleanup in snd_hdac_bus_alloc_stream_pages()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
sound/hda/core/controller.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/sound/hda/core/controller.c b/sound/hda/core/controller.c
index a7acd3418ae8a..32574dc899fad 100644
--- a/sound/hda/core/controller.c
+++ b/sound/hda/core/controller.c
@@ -764,10 +764,13 @@ int snd_hdac_bus_alloc_stream_pages(struct hdac_bus *bus)
error_posbuf:
snd_dma_free_pages(&bus->posbuf);
+ bus->posbuf.area = NULL;
error_bdl:
list_for_each_entry(s, &bus->stream_list, list) {
- if (s->bdl.area)
+ if (s->bdl.area) {
snd_dma_free_pages(&s->bdl);
+ s->bdl.area = NULL;
+ }
}
return -ENOMEM;
}
--
2.53.0
reply other threads:[~2026-10-28 14:44 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261028144443.992569-1-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sound@vger.kernel.org \
--cc=perex@perex.cz \
--cc=tiwai@suse.com \
--cc=zhaodongdong@kylinos.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®