stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Jia-Ju Bai <baijiaju1990@163.com>,
	Takashi Iwai <tiwai@suse.de>
Subject: [PATCH 4.13 22/53] ALSA: seq: Fix copy_from_user() call inside lock
Date: Mon, 16 Oct 2017 18:16:19 +0200	[thread overview]
Message-ID: <20171016161443.216168120@linuxfoundation.org> (raw)
In-Reply-To: <20171016161442.263947886@linuxfoundation.org>

4.13-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Takashi Iwai <tiwai@suse.de>

commit 5803b023881857db32ffefa0d269c90280a67ee0 upstream.

The event handler in the virmidi sequencer code takes a read-lock for
the linked list traverse, while it's calling snd_seq_dump_var_event()
in the loop.  The latter function may expand the user-space data
depending on the event type.  It eventually invokes copy_from_user(),
which might be a potential dead-lock.

The sequencer core guarantees that the user-space data is passed only
with atomic=0 argument, but snd_virmidi_dev_receive_event() ignores it
and always takes read-lock().  For avoiding the problem above, this
patch introduces rwsem for non-atomic case, while keeping rwlock for
atomic case.

Also while we're at it: the superfluous irq flags is dropped in
snd_virmidi_input_open().

Reported-by: Jia-Ju Bai <baijiaju1990@163.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 include/sound/seq_virmidi.h  |    1 +
 sound/core/seq/seq_virmidi.c |   27 +++++++++++++++++++--------
 2 files changed, 20 insertions(+), 8 deletions(-)

--- a/include/sound/seq_virmidi.h
+++ b/include/sound/seq_virmidi.h
@@ -60,6 +60,7 @@ struct snd_virmidi_dev {
 	int port;			/* created/attached port */
 	unsigned int flags;		/* SNDRV_VIRMIDI_* */
 	rwlock_t filelist_lock;
+	struct rw_semaphore filelist_sem;
 	struct list_head filelist;
 };
 
--- a/sound/core/seq/seq_virmidi.c
+++ b/sound/core/seq/seq_virmidi.c
@@ -77,13 +77,17 @@ static void snd_virmidi_init_event(struc
  * decode input event and put to read buffer of each opened file
  */
 static int snd_virmidi_dev_receive_event(struct snd_virmidi_dev *rdev,
-					 struct snd_seq_event *ev)
+					 struct snd_seq_event *ev,
+					 bool atomic)
 {
 	struct snd_virmidi *vmidi;
 	unsigned char msg[4];
 	int len;
 
-	read_lock(&rdev->filelist_lock);
+	if (atomic)
+		read_lock(&rdev->filelist_lock);
+	else
+		down_read(&rdev->filelist_sem);
 	list_for_each_entry(vmidi, &rdev->filelist, list) {
 		if (!vmidi->trigger)
 			continue;
@@ -97,7 +101,10 @@ static int snd_virmidi_dev_receive_event
 				snd_rawmidi_receive(vmidi->substream, msg, len);
 		}
 	}
-	read_unlock(&rdev->filelist_lock);
+	if (atomic)
+		read_unlock(&rdev->filelist_lock);
+	else
+		up_read(&rdev->filelist_sem);
 
 	return 0;
 }
@@ -115,7 +122,7 @@ int snd_virmidi_receive(struct snd_rawmi
 	struct snd_virmidi_dev *rdev;
 
 	rdev = rmidi->private_data;
-	return snd_virmidi_dev_receive_event(rdev, ev);
+	return snd_virmidi_dev_receive_event(rdev, ev, true);
 }
 #endif  /*  0  */
 
@@ -130,7 +137,7 @@ static int snd_virmidi_event_input(struc
 	rdev = private_data;
 	if (!(rdev->flags & SNDRV_VIRMIDI_USE))
 		return 0; /* ignored */
-	return snd_virmidi_dev_receive_event(rdev, ev);
+	return snd_virmidi_dev_receive_event(rdev, ev, atomic);
 }
 
 /*
@@ -209,7 +216,6 @@ static int snd_virmidi_input_open(struct
 	struct snd_virmidi_dev *rdev = substream->rmidi->private_data;
 	struct snd_rawmidi_runtime *runtime = substream->runtime;
 	struct snd_virmidi *vmidi;
-	unsigned long flags;
 
 	vmidi = kzalloc(sizeof(*vmidi), GFP_KERNEL);
 	if (vmidi == NULL)
@@ -223,9 +229,11 @@ static int snd_virmidi_input_open(struct
 	vmidi->client = rdev->client;
 	vmidi->port = rdev->port;	
 	runtime->private_data = vmidi;
-	write_lock_irqsave(&rdev->filelist_lock, flags);
+	down_write(&rdev->filelist_sem);
+	write_lock_irq(&rdev->filelist_lock);
 	list_add_tail(&vmidi->list, &rdev->filelist);
-	write_unlock_irqrestore(&rdev->filelist_lock, flags);
+	write_unlock_irq(&rdev->filelist_lock);
+	up_write(&rdev->filelist_sem);
 	vmidi->rdev = rdev;
 	return 0;
 }
@@ -264,9 +272,11 @@ static int snd_virmidi_input_close(struc
 	struct snd_virmidi_dev *rdev = substream->rmidi->private_data;
 	struct snd_virmidi *vmidi = substream->runtime->private_data;
 
+	down_write(&rdev->filelist_sem);
 	write_lock_irq(&rdev->filelist_lock);
 	list_del(&vmidi->list);
 	write_unlock_irq(&rdev->filelist_lock);
+	up_write(&rdev->filelist_sem);
 	snd_midi_event_free(vmidi->parser);
 	substream->runtime->private_data = NULL;
 	kfree(vmidi);
@@ -520,6 +530,7 @@ int snd_virmidi_new(struct snd_card *car
 	rdev->rmidi = rmidi;
 	rdev->device = device;
 	rdev->client = -1;
+	init_rwsem(&rdev->filelist_sem);
 	rwlock_init(&rdev->filelist_lock);
 	INIT_LIST_HEAD(&rdev->filelist);
 	rdev->seq_mode = SNDRV_VIRMIDI_SEQ_DISPATCH;

  parent reply	other threads:[~2017-10-16 16:19 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-16 16:15 [PATCH 4.13 00/53] 4.13.8-stable review Greg Kroah-Hartman
2017-10-16 16:15 ` [PATCH 4.13 01/53] USB: dummy-hcd: Fix deadlock caused by disconnect detection Greg Kroah-Hartman
2017-10-16 16:15 ` [PATCH 4.13 02/53] MIPS: math-emu: Remove pr_err() calls from fpu_emu() Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 03/53] MIPS: bpf: Fix uninitialised target compiler error Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 04/53] mei: always use domain runtime pm callbacks Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 05/53] dmaengine: edma: Align the memcpy acnt array size with the transfer Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 06/53] dmaengine: ti-dma-crossbar: Fix possible race condition with dma_inuse Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 07/53] NFS: Fix uninitialized rpc_wait_queue Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 08/53] nfs/filelayout: fix oops when freeing filelayout segment Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 09/53] HID: usbhid: fix out-of-bounds bug Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 12/53] KVM: MMU: always terminate page walks at level 1 Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 13/53] KVM: nVMX: fix guest CR4 loading when emulating L2 to L1 exit Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 14/53] usb: renesas_usbhs: Fix DMAC sequence for receiving zero-length packet Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 16/53] iommu/amd: Finish TLB flush in amd_iommu_unmap() Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 17/53] device property: Track owner device of device property Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 18/53] Revert "vmalloc: back off when the current task is killed" Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 19/53] fs/mpage.c: fix mpage_writepage() for pages with buffers Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 20/53] ALSA: usb-audio: Kill stray URB at exiting Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 21/53] ALSA: seq: Fix use-after-free at creating a port Greg Kroah-Hartman
2017-10-16 16:16 ` Greg Kroah-Hartman [this message]
2017-10-16 16:16 ` [PATCH 4.13 23/53] ALSA: caiaq: Fix stray URB at probe error path Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 24/53] ALSA: line6: Fix NULL dereference at podhd_disconnect() Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 25/53] ALSA: line6: Fix missing initialization before error path Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 26/53] ALSA: line6: Fix leftover URB at error-path during probe Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 27/53] drm/atomic: Unref duplicated drm_atomic_state in drm_atomic_helper_resume() Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 28/53] drm/i915/edp: Get the Panel Power Off timestamp after panel is off Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 31/53] drm/i915: Use crtc_state_is_legacy_gamma in intel_color_check Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 32/53] usb: gadget: configfs: Fix memory leak of interface directory data Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 33/53] usb: gadget: composite: Fix use-after-free in usb_composite_overwrite_options Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 34/53] PCI: aardvark: Move to struct pci_host_bridge IRQ mapping functions Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 35/53] Revert "PCI: tegra: Do not allocate MSI target memory" Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 36/53] direct-io: Prevent NULL pointer access in submit_page_section Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 37/53] fix unbalanced page refcounting in bio_map_user_iov Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 38/53] more bio_map_user_iov() leak fixes Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 39/53] bio_copy_user_iov(): dont ignore ->iov_offset Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 40/53] perf script: Add missing separator for "-F ip,brstack" (and brstackoff) Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 41/53] genirq/cpuhotplug: Enforce affinity setting on startup of managed irqs Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 42/53] genirq/cpuhotplug: Add sanity check for effective affinity mask Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 43/53] USB: serial: ftdi_sio: add id for Cypress WICED dev board Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 44/53] USB: serial: cp210x: fix partnum regression Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 45/53] USB: serial: cp210x: add support for ELV TFD500 Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 46/53] USB: serial: option: add support for TP-Link LTE module Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 47/53] USB: serial: qcserial: add Dell DW5818, DW5819 Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 48/53] USB: serial: console: fix use-after-free on disconnect Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 49/53] USB: serial: console: fix use-after-free after failed setup Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 50/53] RAS/CEC: Use the right length for "cec_disable" Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 51/53] x86/microcode: Do the family check first Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 52/53] x86/alternatives: Fix alt_max_short macro to really be a max() Greg Kroah-Hartman
2017-10-16 16:16 ` [PATCH 4.13 53/53] KVM: nVMX: update last_nonleaf_level when initializing nested EPT Greg Kroah-Hartman
2017-10-16 23:41 ` [PATCH 4.13 00/53] 4.13.8-stable review Shuah Khan
2017-10-17  6:59   ` Greg Kroah-Hartman
2017-10-17  0:25 ` Guenter Roeck
2017-10-17 13:21   ` Greg Kroah-Hartman
     [not found] ` <20171016161443.534299546@linuxfoundation.org>
     [not found]   ` <866e97b1-08dd-fd43-7713-699759f63fcf@3CityElectronics.com>
2017-10-17  7:02     ` [PATCH 4.13 30/53] drm/i915/bios: parse DDI ports also for CHV for HDMI DDC pin and DP AUX channel Jani Nikula

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=20171016161443.216168120@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=baijiaju1990@163.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tiwai@suse.de \
    /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;
as well as URLs for NNTP newsgroup(s).