From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Takashi Iwai <tiwai@suse.de>,
Liam Girdwood <liam.r.girdwood@linux.intel.com>,
Mark Brown <broonie@kernel.org>
Subject: [PATCH 3.14 27/73] ASoC: dpcm: Fix race between FE/BE updates and trigger
Date: Fri, 5 Dec 2014 14:44:33 -0800 [thread overview]
Message-ID: <20141205224434.759994796@linuxfoundation.org> (raw)
In-Reply-To: <20141205224433.921659956@linuxfoundation.org>
3.14-stable review patch. If anyone has any objections, please let me know.
------------------
From: Takashi Iwai <tiwai@suse.de>
commit ea9d0d771fcd32cd56070819749477d511ec9117 upstream.
DPCM can update the FE/BE connection states totally asynchronously
from the FE's PCM state. Most of FE/BE state changes are protected by
mutex, so that they won't race, but there are still some actions that
are uncovered. For example, suppose to switch a BE while a FE's
stream is running. This would call soc_dpcm_runtime_update(), which
sets FE's runtime_update flag, then sets up and starts BEs, and clears
FE's runtime_update flag again.
When a device emits XRUN during this operation, the PCM core triggers
snd_pcm_stop(XRUN). Since the trigger action is an atomic ops, this
isn't blocked by the mutex, thus it kicks off DPCM's trigger action.
It eventually updates and clears FE's runtime_update flag while
soc_dpcm_runtime_update() is running concurrently, and it results in
confusion.
Usually, for avoiding such a race, we take a lock. There is a PCM
stream lock for that purpose. However, as already mentioned, the
trigger action is atomic, and we can't take the lock for the whole
soc_dpcm_runtime_update() or other operations that include the lengthy
jobs like hw_params or prepare.
This patch provides an alternative solution. This adds a way to defer
the conflicting trigger callback to be executed at the end of FE/BE
state changes. For doing it, two things are introduced:
- Each runtime_update state change of FEs is protected via PCM stream
lock.
- The FE's trigger callback checks the runtime_update flag. If it's
not set, the trigger action is executed there. If set, mark the
pending trigger action and returns immediately.
- At the exit of runtime_update state change, it checks whether the
pending trigger is present. If yes, it executes the trigger action
at this point.
Reported-and-tested-by: Qiao Zhou <zhouqiao@marvell.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Acked-by: Liam Girdwood <liam.r.girdwood@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
include/sound/soc-dpcm.h | 2 +
sound/soc/soc-pcm.c | 72 ++++++++++++++++++++++++++++++++++++-----------
2 files changed, 58 insertions(+), 16 deletions(-)
--- a/include/sound/soc-dpcm.h
+++ b/include/sound/soc-dpcm.h
@@ -102,6 +102,8 @@ struct snd_soc_dpcm_runtime {
/* state and update */
enum snd_soc_dpcm_update runtime_update;
enum snd_soc_dpcm_state state;
+
+ int trigger_pending; /* trigger cmd + 1 if pending, 0 if not */
};
/* can this BE stop and free */
--- a/sound/soc/soc-pcm.c
+++ b/sound/soc/soc-pcm.c
@@ -1258,13 +1258,36 @@ static void dpcm_set_fe_runtime(struct s
dpcm_init_runtime_hw(runtime, &cpu_dai_drv->capture);
}
+static int dpcm_fe_dai_do_trigger(struct snd_pcm_substream *substream, int cmd);
+
+/* Set FE's runtime_update state; the state is protected via PCM stream lock
+ * for avoiding the race with trigger callback.
+ * If the state is unset and a trigger is pending while the previous operation,
+ * process the pending trigger action here.
+ */
+static void dpcm_set_fe_update_state(struct snd_soc_pcm_runtime *fe,
+ int stream, enum snd_soc_dpcm_update state)
+{
+ struct snd_pcm_substream *substream =
+ snd_soc_dpcm_get_substream(fe, stream);
+
+ snd_pcm_stream_lock_irq(substream);
+ if (state == SND_SOC_DPCM_UPDATE_NO && fe->dpcm[stream].trigger_pending) {
+ dpcm_fe_dai_do_trigger(substream,
+ fe->dpcm[stream].trigger_pending - 1);
+ fe->dpcm[stream].trigger_pending = 0;
+ }
+ fe->dpcm[stream].runtime_update = state;
+ snd_pcm_stream_unlock_irq(substream);
+}
+
static int dpcm_fe_dai_startup(struct snd_pcm_substream *fe_substream)
{
struct snd_soc_pcm_runtime *fe = fe_substream->private_data;
struct snd_pcm_runtime *runtime = fe_substream->runtime;
int stream = fe_substream->stream, ret = 0;
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_FE);
ret = dpcm_be_dai_startup(fe, fe_substream->stream);
if (ret < 0) {
@@ -1286,13 +1309,13 @@ static int dpcm_fe_dai_startup(struct sn
dpcm_set_fe_runtime(fe_substream);
snd_pcm_limit_hw_rates(runtime);
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
return 0;
unwind:
dpcm_be_dai_startup_unwind(fe, fe_substream->stream);
be_err:
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
return ret;
}
@@ -1339,7 +1362,7 @@ static int dpcm_fe_dai_shutdown(struct s
struct snd_soc_pcm_runtime *fe = substream->private_data;
int stream = substream->stream;
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_FE);
/* shutdown the BEs */
dpcm_be_dai_shutdown(fe, substream->stream);
@@ -1353,7 +1376,7 @@ static int dpcm_fe_dai_shutdown(struct s
dpcm_dapm_stream_event(fe, stream, SND_SOC_DAPM_STREAM_STOP);
fe->dpcm[stream].state = SND_SOC_DPCM_STATE_CLOSE;
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
return 0;
}
@@ -1401,7 +1424,7 @@ static int dpcm_fe_dai_hw_free(struct sn
int err, stream = substream->stream;
mutex_lock_nested(&fe->card->mutex, SND_SOC_CARD_CLASS_RUNTIME);
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_FE);
dev_dbg(fe->dev, "ASoC: hw_free FE %s\n", fe->dai_link->name);
@@ -1416,7 +1439,7 @@ static int dpcm_fe_dai_hw_free(struct sn
err = dpcm_be_dai_hw_free(fe, stream);
fe->dpcm[stream].state = SND_SOC_DPCM_STATE_HW_FREE;
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
mutex_unlock(&fe->card->mutex);
return 0;
@@ -1509,7 +1532,7 @@ static int dpcm_fe_dai_hw_params(struct
int ret, stream = substream->stream;
mutex_lock_nested(&fe->card->mutex, SND_SOC_CARD_CLASS_RUNTIME);
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_FE);
memcpy(&fe->dpcm[substream->stream].hw_params, params,
sizeof(struct snd_pcm_hw_params));
@@ -1532,7 +1555,7 @@ static int dpcm_fe_dai_hw_params(struct
fe->dpcm[stream].state = SND_SOC_DPCM_STATE_HW_PARAMS;
out:
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
mutex_unlock(&fe->card->mutex);
return ret;
}
@@ -1646,7 +1669,7 @@ int dpcm_be_dai_trigger(struct snd_soc_p
}
EXPORT_SYMBOL_GPL(dpcm_be_dai_trigger);
-static int dpcm_fe_dai_trigger(struct snd_pcm_substream *substream, int cmd)
+static int dpcm_fe_dai_do_trigger(struct snd_pcm_substream *substream, int cmd)
{
struct snd_soc_pcm_runtime *fe = substream->private_data;
int stream = substream->stream, ret;
@@ -1720,6 +1743,23 @@ out:
return ret;
}
+static int dpcm_fe_dai_trigger(struct snd_pcm_substream *substream, int cmd)
+{
+ struct snd_soc_pcm_runtime *fe = substream->private_data;
+ int stream = substream->stream;
+
+ /* if FE's runtime_update is already set, we're in race;
+ * process this trigger later at exit
+ */
+ if (fe->dpcm[stream].runtime_update != SND_SOC_DPCM_UPDATE_NO) {
+ fe->dpcm[stream].trigger_pending = cmd + 1;
+ return 0; /* delayed, assuming it's successful */
+ }
+
+ /* we're alone, let's trigger */
+ return dpcm_fe_dai_do_trigger(substream, cmd);
+}
+
int dpcm_be_dai_prepare(struct snd_soc_pcm_runtime *fe, int stream)
{
struct snd_soc_dpcm *dpcm;
@@ -1763,7 +1803,7 @@ static int dpcm_fe_dai_prepare(struct sn
dev_dbg(fe->dev, "ASoC: prepare FE %s\n", fe->dai_link->name);
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_FE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_FE);
/* there is no point preparing this FE if there are no BEs */
if (list_empty(&fe->dpcm[stream].be_clients)) {
@@ -1790,7 +1830,7 @@ static int dpcm_fe_dai_prepare(struct sn
fe->dpcm[stream].state = SND_SOC_DPCM_STATE_PREPARE;
out:
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
mutex_unlock(&fe->card->mutex);
return ret;
@@ -1937,11 +1977,11 @@ static int dpcm_run_new_update(struct sn
{
int ret;
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_BE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_BE);
ret = dpcm_run_update_startup(fe, stream);
if (ret < 0)
dev_err(fe->dev, "ASoC: failed to startup some BEs\n");
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
return ret;
}
@@ -1950,11 +1990,11 @@ static int dpcm_run_old_update(struct sn
{
int ret;
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_BE;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_BE);
ret = dpcm_run_update_shutdown(fe, stream);
if (ret < 0)
dev_err(fe->dev, "ASoC: failed to shutdown some BEs\n");
- fe->dpcm[stream].runtime_update = SND_SOC_DPCM_UPDATE_NO;
+ dpcm_set_fe_update_state(fe, stream, SND_SOC_DPCM_UPDATE_NO);
return ret;
}
next prev parent reply other threads:[~2014-12-05 22:44 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-12-05 22:44 [PATCH 3.14 00/73] 3.14.26-stable review Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 01/73] MIPS: oprofile: Fix backtrace on 64-bit kernel Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 02/73] MIPS: Loongson: Make platform serial setup always built-in Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 03/73] x86_64, traps: Fix the espfix64 #DF fixup and rewrite it in C Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 04/73] x86_64, traps: Stop using IST for #SS Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 05/73] x86_64, traps: Rework bad_iret Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 06/73] x86: Require exact match for noxsave command line option Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 07/73] x86, mm: Set NX across entire PMD at boot Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 08/73] x86, kaslr: Handle Gold linker for finding bss/brk Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 09/73] uprobes, x86: Fix _TIF_UPROBE vs _TIF_NOTIFY_RESUME Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 10/73] sparc64: Fix constraints on swab helpers Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 11/73] inetdevice: fixed signed integer overflow Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 12/73] ipv4: Fix incorrect error code when adding an unreachable route Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 13/73] ieee802154: fix error handling in ieee802154fake_probe() Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 15/73] bonding: fix curr_active_slave/carrier with loadbalance arp monitoring Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 16/73] pptp: fix stack info leak in pptp_getname() Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 17/73] ipx: fix locking regression in ipx_sendmsg and ipx_recvmsg Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 18/73] PCI: Support 64-bit bridge windows if we have 64-bit dma_addr_t Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 19/73] PCI/MSI: Add device flag indicating that 64-bit MSIs dont work Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 20/73] clockevent: sun4i: Fix race condition in the probe code Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 21/73] IB/isert: Adjust CQ size to HW limits Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 22/73] ib_isert: Add max_send_sge=2 minimum for control PDU responses Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 23/73] ASoC: rsnd: remove unsupported PAUSE flag Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 24/73] ASoC: fsi: " Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 25/73] ASoC: sgtl5000: Fix SMALL_POP bit definition Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 26/73] ASoC: wm_adsp: Avoid attempt to free buffers that might still be in use Greg Kroah-Hartman
2014-12-05 22:44 ` Greg Kroah-Hartman [this message]
2014-12-05 22:44 ` [PATCH 3.14 28/73] ath9k: Fix RTC_DERIVED_CLK usage Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 29/73] of/base: Fix PowerPC address parsing hack Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 30/73] powerpc/pseries: Honor the generic "no_64bit_msi" flag Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 31/73] powerpc/pseries: Fix endiannes issue in RTAS call from xmon Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 32/73] iio: Fix IIO_EVENT_CODE_EXTRACT_DIR bit mask Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 33/73] staging: r8188eu: Add new device ID for DLink GO-USB-N150 Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 34/73] USB: ssu100: fix overrun-error reporting Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 35/73] USB: keyspan: " Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 36/73] USB: keyspan: fix tty line-status reporting Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 37/73] USB: serial: cp210x: add IDs for CEL MeshConnect USB Stick Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 38/73] usb: serial: ftdi_sio: add PIDs for Matrix Orbital products Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 39/73] usb-quirks: Add reset-resume quirk for MS Wireless Laser Mouse 6000 Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 40/73] USB: xhci: dont start a halted endpoint before its new dequeue is set Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 41/73] usb: xhci: rework root port wake bits if controller isnt allowed to wakeup Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 42/73] can: esd_usb2: fix memory leak on disconnect Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 43/73] ALSA: usb-audio: Add ctrl message delay quirk for Marantz/Denon devices Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 45/73] of/irq: Drop obsolete interrupts vs interrupts-extended text Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 46/73] ARM: 8216/1: xscale: correct auxiliary register in suspend/resume Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 47/73] ARM: 8222/1: mvebu: enable strex backoff delay Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 48/73] ARM: 8226/1: cacheflush: get rid of restarting block Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 49/73] Input: synaptics - adjust min/max on Thinkpad E540 Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 50/73] Input: xpad - use proper endpoint type Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 51/73] srp-target: Retry when QP creation fails with ENOMEM Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 52/73] target: Dont call TFO->write_pending if data_length == 0 Greg Kroah-Hartman
2014-12-05 22:44 ` [PATCH 3.14 53/73] iser-target: Handle DEVICE_REMOVAL event on network portal listener correctly Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 54/73] Target/iser: Fix a wrong dereference in case discovery session is over iser Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 55/73] iser-target: Disable TX completion interrupt coalescing Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 56/73] spi: dw: Fix dynamic speed change Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 57/73] vhost-scsi: Take configfs group dependency during VHOST_SCSI_SET_ENDPOINT Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 60/73] rt2x00: do not align payload on modern H/W Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 61/73] nfsd: correctly define v4.2 support attributes Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 62/73] nfsd: Fix slot wake up race in the nfsv4.1 callback code Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 63/73] nfsd: dont halt scanning the DRC LRU list when theres an RC_INPROG entry Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 64/73] net/ping: handle protocol mismatching scenario Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 65/73] bnx2fc: do not add shared skbs to the fcoe_rx_list Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 66/73] drm/radeon: fix endian swapping in vbios fetch for tdp table Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 67/73] gpu/radeon: Set flag to indicate broken 64-bit MSI Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 68/73] bitops: Fix shift overflow in GENMASK macros Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 69/73] locks: eliminate BUG() call when theres an unexpected lock on file close Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 70/73] powerpc/powernv: Honor the generic "no_64bit_msi" flag Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 71/73] batman: fix a bogus warning from batadv_is_on_batman_iface() Greg Kroah-Hartman
2014-12-05 22:45 ` [PATCH 3.14 72/73] x86: kvm: use alternatives for VMCALL vs. VMMCALL if kernel text is read-only Greg Kroah-Hartman
2014-12-06 3:30 ` [PATCH 3.14 00/73] 3.14.26-stable review Guenter Roeck
2014-12-06 21:36 ` Shuah Khan
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=20141205224434.759994796@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=broonie@kernel.org \
--cc=liam.r.girdwood@linux.intel.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).