ALSA: usb-audio: Avoid unnecessary EP setups in prepare
The recent fix for USB suspend breakage moved the code to set up EP
from hw_params to prepare, but it means also the EP setup might be
called multiple times unnecessarily because the prepare callback can
be called multiple times without starting the stream (e.g. OSS
emulation).
This patch adds a new flag to struct snd_usb_substream indicating
whether the setup of EP is required, and do it only when necessary,
i.e. right after hw_params or suspend.
Move interface and endpoint configuration from hw_params to prepare
callback. During system suspend/resume when the USB device power isn't
cycled the interface and endpoint configuration need to be set before
audio playback can continue. Resume involves another call to prepare
but not to hw_params, moving it here allows a playing stream to continue
after resume.
ALSA: usb-audio: Don't require hw_params in endpoint.
Change the interface to configure an endpoint so that it doesn't require
a hw_params struct. This will allow it to be called from prepare
instead of hw_params, configuring it after system resume.
ALSA: hda - limit internal mic boost for Asus X202E
When the input gain for the internal mic is set to its maximum level,
the background noise becomes so high - and any relevant signal clipped -
that the setting becomes unusable. It is better to limit the amplification.
Commit ALSA: compress_core: integer overflow in snd_compr_allocate_buffer()
added a new error check for input params.
this add new routine for input checks and moves buffer overflow check to this
new routine. This allows the error value to be propogated to user space
ALSA: hda - Allow to pass position_fix=0 explicitly
Set the default value of position_fix -1, and allow user passing
position_fix=0 explicitly to set the "auto" position-fix mode.
Otherwise the auto mode may be switched to others like COMBO of
VIACOMBO when the controller prefers it, thus user can't set the auto
mode any longer.
Also updated the documentation appropriately, too.
ALSA: hda - Fix disordered enum definitions in patch_cirrus.c
Due to the definitions of CS420X_IMAC27_122 and CS420X_APPLE as
aliases, the rest enums are set to duplicated values unexpectedly.
Move the alias definitions at the end so that the enum values are
defined in the proper order.
For following the standard, define more channel map positions and
shuffle the items a bit:
- As both PulseAudio and gstreamer define MONO channel position
explicitly, we should follow that, too. The mono streams point to
this channel position unless they are explicitly assigned to certain
channel positions.
- Top-front-* and Top-rear-* positions are added, carried from
PulseAudio's definitions.
- Move NA and MONO definitions at the top of table right after
UNKNOWN, since these are more abstract in comparison with other
practical positions.
Provide channel maps for individual stereo streams of ENS1370 and
ENS1371. Note that the configuration of ENS1370 uses the secondary
PCM as the front unlike ENS1371.
MacBook Pro 10,1 needs a few adjustments to make it working:
- more COEF verbs
- some pin config overrides to disable the unused pin (0x0d, 0x12),
and fix the internal mic (0x0e)
In addition, it uses GPIO 1 and 3 like other MacBooks.
The internal digital mic on the machine is still problematic: it seems
that only the right channel is used and the left is always static.
This looks like a hardware design, so we need to cope in the software
side somehow...
The primary information and test were brought from Daniel J Blueman.
ALSA: hda - Avoid BDL position workaround when no_period_wakeup is set
Originally the bogus period at BDL head was introduced as a workaround
for the mismatching position update at the period boundary, typically
seen on dmix. However, for applications like PulseAudio that don't
require period wake ups, this workaround is just superfluous. Thus
better to disable it when no_period_wakeup is given in hw_params.
ALSA: hda_intel: add position_fix quirk for Asus K53E
Commit c20c5a841cbe47f5b7812b57bd25397497e5fbc0 changed some chipsets to
default to POS_FIX_COMBO so they now use POS_FIX_LPIB instead of
POS_FIX_POSBUF. Since then I've been getting artifacts on playback, including
repeated sounds on my Asus laptop.
My hardware is Cougar Point which the commit log of c20c5a841cbe47f5b7812b57bd25397497e5fbc0 mentions as tested so POS_FIX_COMBO
probably works in general but apparently it doesn't on Asus K53E therefore the
need for the quirk.
Dan Carpenter [Tue, 11 Sep 2012 11:12:43 +0000 (14:12 +0300)]
ALSA: compress_core: fix open flags test in snd_compr_open()
O_RDONLY is zero so the original test (f->f_flags & O_RDONLY) is always
false and it will never do compress capture. The test for O_WRONLY is
also slightly off. The original test would consider "->flags =
(O_WRONLY | O_RDWR)" as write only instead of rejecting it as invalid.
I've also removed the pr_err() because that could flood dmesg.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
snd_hda_codec_reset() calls restore_pincfgs() where the codec is
powered up again, which eventually tries to resume and initialize via
the callbacks of the codec. However, it's the place just after codec
free callback, thus no codec callbacks should be called after that.
On a codec like CS4206, it results in Oops due to the access in init
callback.
This patch fixes the issue by clearing the codec callbacks properly
after freeing codec.
The purpose of this flag is unclear. If the problem is that some machines
have broken misc/NO_PRESENCE bits, they should be fixed by pin fixups.
In addition, this causes jack detection functionality to be flawed on
the M31EI, where there are two jacks without jack detection (which is
properly marked as NO_PRESENCE), but due to ignore_misc_bit, these
jacks are instead being reported as being present but always unplugged.
BugLink: https://bugs.launchpad.net/bugs/939161 Signed-off-by: David Henningsson <david.henningsson@canonical.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Iwai [Tue, 21 Aug 2012 12:47:18 +0000 (14:47 +0200)]
ALSA: Follow channel position definitions to alsa-lib mixer
There is already a set of channel position definitions in alsa-lib
mixer.h, and it'd be more practical to keep the same order for the
PCM channel map, too. The value is shifted with 1 to keep zero for
UNKNOWN.
Takashi Iwai [Tue, 21 Aug 2012 12:49:47 +0000 (14:49 +0200)]
ALSA: Remove VOLATILE flag from chmap ctls
The VOLATILE flag was added to control elements by
snd_pcm_add_chmap_ctls() just because I didn't want to have a
side-effect of "alsactl restore". But now the set operation doesn't
allow to change the value unless the PCM stream is in PREAPRED state,
there is no reason to keep this flag. Let's rip it off.
Takashi Iwai [Fri, 3 Aug 2012 15:59:36 +0000 (17:59 +0200)]
ALSA: ac97: Implement channel map workaround for ALC650
ALC650 has a channel swap option between surround and CLFE channels,
so we need to tweak the channel maps dynamically depending on the
register bit.
Now struct snd_ac97 can contain chmap pointers for playback and
capture. The driver may store these and let ac97 driver changing the
channel mapping dynamically.
ALSA: hda - Provide the proper channel mapping for generic HDMI driver
... instead of the standard fixed channel maps.
The generic HDMI is based on the audio infoframe, and its configuration
can be selected via CA bits. Thus we need a translation between the
CA index and the verbose channel map list.
Although HD-audio allows pair-wise channel configurations, only the
fixed channel positions are used in this version. In future, this can
be changed and allow user to modify the channel positions.
This patch implements the basic data types for the standard channel
mapping API handling.
- The definitions of the channel positions and the new TLV types are
added in sound/asound.h and sound/tlv.h, so that they can be
referred from user-space.
- Introduced a new helper function snd_pcm_add_chmap_ctls() to create
control elements representing the channel maps for each PCM
(sub)stream.
- Some standard pre-defined channel maps are provided for
convenience.
ALSA: hda - Move non-PCM check to per_pin in patch_hdmi.c
Recently the check for non-PCM stream state was added to the generic
HDMI driver code. But this check should be done rather to each pin
instead of each converter. Otherwise when a different converter is
assigned at the next open, the audio infoframe can be inconsistent
with the setup using the previous converter.
For fixing this issue, this patch moves the state of the current
non-PCM status from per_cvt to per_pin. (In addition an unused
argument cvt_nid is stripped from hdmi_setup_channel_mapping())
ALSA: usb-audio: Fix bogus error messages for delay accounting
The recent fix for the missing fine delayed time adjustment gives
strange error messages at each start of the playback stream, such as
delay: estimated 0, actual 352
delay: estimated 353, actual 705
These come from the sanity check in retire_playback_urb(). Before the
stream is activated via start_endpoints(), a few silent packets have
been already sent. And at this point the delay account is still in
the state as if the new packets are just queued, so the driver gets
confused and spews the bogus error messages.
For fixing the issue, we just need to check whether the received
packet is valid, whether it's zero sized or not.
ALSA: hda - fix control names for multiple speaker out on IDT/STAC
For multiple speaker outs, the names were previously
"Speaker,0", "Speaker,1", "Center"/"LFE", "Speaker,3". This is
inconsistent, confusing, and is not picked up correctly by PulseAudio.
Instead use "Front", "Surround", "Center"/"LFE", "Side" which
is more standard.
ALSA: hda - Fix missing Master volume for STAC9200/925x
With the commit [2faa3bf: ALSA: hda - Rewrite the mute-LED hook with
vmaster hook in patch_sigmatel.c], the former Master volume control
was converted to PCM. This was supposed to be covered by the vmaster
control. But due to the lack of "PCM" slave definition, this didn't
happen properly. The patch fixes the missing entry.
Wang Xingchao [Thu, 6 Sep 2012 02:02:38 +0000 (10:02 +0800)]
ALSA: HDMI - Setup channel mapping for non_pcm audio
For HBR stream test, use straight channel mapping way.
when switched back to "speaker-test -c8", even the audio
infoframe is up-to-date, there should be correct channel mapping setup.
Signed-off-by: Wang Xingchao <xingchao.wang@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Wang Xingchao [Thu, 6 Sep 2012 02:02:37 +0000 (10:02 +0800)]
ALSA: HDMI - Enable HBR feature on Intel chips
HDMI channel remapping apparently effects HBR packets on Intel's chips.
For compressed non-PCM audio, use "straight-through" channel mapping.
For uncompressed multi-channel pcm audio, use normal channel mapping.
Signed-off-by: Wang Xingchao <xingchao.wang@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Jaroslav Kysela [Tue, 4 Sep 2012 09:21:45 +0000 (11:21 +0200)]
ALSA: remove the main version information
Remove the main ALSA version number from the kernel ALSA driver.
The ALSA driver package release diverges from the upstream. This may
confuse users to see the same ALSA version for many kernel releases
and this version lost it's original purpose and connection.
The "ioctl" APIs have own version numbers, so the user space may check
for specific API changes only.
Daniel Mack [Tue, 4 Sep 2012 08:23:07 +0000 (10:23 +0200)]
ALSA: snd-usb: Add quirks for Playback Designs devices
Playback Designs' USB devices have some hardware limitations on their
USB interface. In particular:
- They need a 20ms delay after each class compliant request as the
hardware ACKs the USB packets before the device is actually ready
for the next command. Sending data immediately will result in buffer
overflows in the hardware.
- The devices send bogus feedback data at the start of each stream
which confuse the feedback format auto-detection.
This patch introduces a new quirks hook that is called after each
control packet and which adds a delay for all devices that match
Playback Designs' USB VID for now.
In addition, it adds a counter to snd_usb_endpoint to drop received
packets on the floor. Another new quirks function that is called once
an endpoint is started initializes that counter for these devices on
their sync endpoint.
Signed-off-by: Daniel Mack <zonque@gmail.com> Reported-and-tested-by: Andreas Koch <andreas@akdesigninc.com> Supported-by: Demian Martin <demianm_1@yahoo.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
ALSA: Remove the last mention of SNDRV_MAIN_OBJECT_FILE
SNDRV_MAIN_OBJECT_FILE hasn't done anything since the pre-git days, and
the only remaining reference occurs as a #define in sound/last.c. Drop
that last mention of it.
Commit 68e67f40b ("ALSA: snd-usb: move calls to usb_set_interface")
saved us some unnecessary calls to snd_usb_set_interface() but ignored
the fact that there is at least one device out there which operates on
two endpoint in different interfaces simultaniously.
Take care for this by catching the case where data and sync endpoints
are located on different interfaces and calling snd_usb_set_interface()
between the start of the two endpoints.
Signed-off-by: Daniel Mack <zonque@gmail.com> Reported-by: Robert M. Albrecht <linux@romal.de> Cc: stable@kernel.org [v3.5+] Signed-off-by: Takashi Iwai <tiwai@suse.de>
Daniel Mack [Thu, 30 Aug 2012 16:52:30 +0000 (18:52 +0200)]
ALSA: snd-usb: fix calls to next_packet_size
In order to support devices with implicit feedback streaming models,
packet sizes are now stored with each individual urb, and the PCM
handling code which fills the buffers purely relies on the size fields
now.
However, calling snd_usb_audio_next_packet_size() for all possible
packets in an URB at once, prior to letting the PCM code do its job
does in fact not lead to the same behaviour than what the old code did:
The PCM code will break its loop once a period boundary is reached,
consequently using up less packets that it really could.
As snd_usb_audio_next_packet_size() implements a feedback mechanism to
the endpoints phase accumulator, the number of calls to that function
matters, and when called too often, the data rate runs out of bounds.
Fix this by making the next_packet function public, and call it from the
PCM code as before if the packet data sizes are not defined.
Daniel Mack [Thu, 30 Aug 2012 16:52:29 +0000 (18:52 +0200)]
ALSA: snd-usb: restore delay information
Parts of commit 294c4fb8 ("ALSA: usb: refine delay information with USB
frame counter") were unfortunately lost during the refactoring of the
snd-usb driver in 3.5.
This patch adds them back, restoring the correct delay information
behaviour.
Takashi Iwai [Fri, 31 Aug 2012 14:46:56 +0000 (07:46 -0700)]
ALSA: hda - Yet another fix for D3 stop-clock refcounting
The call of pm_notify callback in snd_hda_codec_free() should be with
the check of the current state whether pm_notify(false) is called or
not, instead of codec->power_on check.
For improving the code readability and fixing this inconsistency,
codec->d3_stop_clk_ok is renamed to codec->pm_down_notified, and this
flag is set only when runtime PM down is called. The new name reflects
to a more direct purpose of the flag.
Takashi Iwai [Fri, 24 Aug 2012 16:38:08 +0000 (18:38 +0200)]
ALSA: hda - Clean up CONFIG_SND_HDA_POWER_SAVE
CONFIG_SND_HDA_POWER_SAVE is no longer an experimental feature and its
behavior can be well controlled via the default value and module
parameter. Let's just replace it with the standard CONFIG_PM.
Takashi Iwai [Tue, 28 Aug 2012 16:59:20 +0000 (09:59 -0700)]
ALSA: hda - Fix D3 clock stop check for codecs with own set_power_state op
When a codec provides its own set_power_state op, the D3-clock-stop
isn't checked correctly. And the recent changes for repeating the
state-setting operation isn't applied to such a codec, too.
This patch fixes these issues by moving the call of codec's own op to
the place where the generic power-set operation is done, and move the
power-state synchronization code out of
snd_hda_set_power_state_to_all() so that it can be called always at
the end of power-up/down sequence, and updates the D3 clock-stop flag
properly.
Takashi Iwai [Tue, 28 Aug 2012 16:14:29 +0000 (09:14 -0700)]
ALSA: hda - Fix runtime PM leftover refcounts
When the HD-audio is removed, it leaves the refcounts when codecs are
powered up (usually yes) in the destructor. For fixing the unbalance,
and cleaning up the code mess, this patch changes the following:
- change pm_notify callback to take the explicit power on/off state,
- check of D3 stop-clock and keep_link_on flags is moved to the caller
side,
- call pm_notify callback in snd_hda_codec_new() and snd_hda_codec_free()
so that the refcounts are proprely updated.
Daniel Mack [Wed, 29 Aug 2012 11:17:05 +0000 (13:17 +0200)]
ALSA: snd-usb: Fix URB cancellation at stream start
Commit e9ba389c5 ("ALSA: usb-audio: Fix scheduling-while-atomic bug in
PCM capture stream") fixed a scheduling-while-atomic bug that happened
when snd_usb_endpoint_start was called from the trigger callback, which
is an atmic context. However, the patch breaks the idea of the endpoints
reference counting, which is the reason why the driver has been
refactored lately.
Revert that commit and let snd_usb_endpoint_start() take care of the URB
cancellation again. As this function is called from both atomic and
non-atomic context, add a flag to denote whether the function may sleep.
Linus Torvalds [Wed, 29 Aug 2012 18:36:22 +0000 (11:36 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
Pull btrfs fixes from Chris Mason:
"I've split out the big send/receive update from my last pull request
and now have just the fixes in my for-linus branch. The send/recv
branch will wander over to linux-next shortly though.
The largest patches in this pull are Josef's patches to fix DIO
locking problems and his patch to fix a crash during balance. They
are both well tested.
The rest are smaller fixes that we've had queued. The last rc came
out while I was hacking new and exciting ways to recover from a
misplaced rm -rf on my dev box, so these missed rc3."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (25 commits)
Btrfs: fix that repair code is spuriously executed for transid failures
Btrfs: fix ordered extent leak when failing to start a transaction
Btrfs: fix a dio write regression
Btrfs: fix deadlock with freeze and sync V2
Btrfs: revert checksum error statistic which can cause a BUG()
Btrfs: remove superblock writing after fatal error
Btrfs: allow delayed refs to be merged
Btrfs: fix enospc problems when deleting a subvol
Btrfs: fix wrong mtime and ctime when creating snapshots
Btrfs: fix race in run_clustered_refs
Btrfs: don't run __tree_mod_log_free_eb on leaves
Btrfs: increase the size of the free space cache
Btrfs: barrier before waitqueue_active
Btrfs: fix deadlock in wait_for_more_refs
btrfs: fix second lock in btrfs_delete_delayed_items()
Btrfs: don't allocate a seperate csums array for direct reads
Btrfs: do not strdup non existent strings
Btrfs: do not use missing devices when showing devname
Btrfs: fix that error value is changed by mistake
Btrfs: lock extents as we map them in DIO
...
David Rientjes [Wed, 29 Aug 2012 02:57:21 +0000 (19:57 -0700)]
mm, slab: lock the correct nodelist after reenabling irqs
cache_grow() can reenable irqs so the cpu (and node) can change, so ensure
that we take list_lock on the correct nodelist.
This fixes an issue with commit 072bb0aa5e06 ("mm: sl[au]b: add
knowledge of PFMEMALLOC reserve pages") where list_lock for the wrong
node was taken after growing the cache.
Reported-and-tested-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: David Rientjes <rientjes@google.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Stefan Behrens [Fri, 10 Aug 2012 14:58:21 +0000 (08:58 -0600)]
Btrfs: fix that repair code is spuriously executed for transid failures
If verify_parent_transid() fails for all mirrors, the current code
calls repair_io_failure() anyway which means:
- that the disk block is rewritten without repairing anything and
- that a kernel log message is printed which misleadingly claims
that a read error was corrected.
This is an example:
parent transid verify failed on 615015833600 wanted 110423 found 110424
parent transid verify failed on 615015833600 wanted 110423 found 110424
btrfs read error corrected: ino 1 off 615015833600 (dev /dev/...)
It is wrong to ignore the results from verify_parent_transid() and to
call repair_eb_io_failure() when the verification of the transids failed.
This commit fixes the issue.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de> Signed-off-by: Chris Mason <chris.mason@oracle.com>
In dio write, we should unlock the section which we didn't do IO on in case that
we fall back to buffered write. But we need to not only unlock the section
but also cleanup reserved space for the section.
This bug was found while running xfstests 133, with this 133 no longer complains.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Josef Bacik [Fri, 24 Aug 2012 18:53:03 +0000 (12:53 -0600)]
Btrfs: fix deadlock with freeze and sync V2
We can deadlock with freeze right now because we unconditionally start a
transaction in our ->sync_fs() call. To fix this just check and see if we
have a running transaction to commit. This saves us from the deadlock
because at this point we'll have the umount sem for the sb so we're safe
from freezes coming in after we've done our check. With this patch the
freeze xfstests no longer deadlocks. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Stefan Behrens [Mon, 27 Aug 2012 14:30:03 +0000 (08:30 -0600)]
Btrfs: revert checksum error statistic which can cause a BUG()
Commit 442a4f6308e694e0fa6025708bd5e4e424bbf51c added btrfs device
statistic counters for detected IO and checksum errors to Linux 3.5.
The statistic part that counts checksum errors in
end_bio_extent_readpage() can cause a BUG() in a subfunction:
"kernel BUG at fs/btrfs/volumes.c:3762!"
That part is reverted with the current patch.
However, the counting of checksum errors in the scrub context remains
active, and the counting of detected IO errors (read, write or flush
errors) in all contexts remains active.
Cc: stable <stable@vger.kernel.org> # 3.5 Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Stefan Behrens [Wed, 1 Aug 2012 11:45:52 +0000 (05:45 -0600)]
Btrfs: remove superblock writing after fatal error
With commit acce952b0, btrfs was changed to flag the filesystem with
BTRFS_SUPER_FLAG_ERROR and switch to read-only mode after a fatal
error happened like a write I/O errors of all mirrors.
In such situations, on unmount, the superblock is written in
btrfs_error_commit_super(). This is done with the intention to be able
to evaluate the error flag on the next mount. A warning is printed
in this case during the next mount and the log tree is ignored.
The issue is that it is possible that the superblock points to a root
that was not written (due to write I/O errors).
The result is that the filesystem cannot be mounted. btrfsck also does
not start and all the other btrfs-progs tools fail to start as well.
However, mount -o recovery is working well and does the right things
to recover the filesystem (i.e., don't use the log root, clear the
free space cache and use the next mountable root that is stored in the
root backup array).
This patch removes the writing of the superblock when
BTRFS_SUPER_FLAG_ERROR is set, and removes the handling of the error
flag in the mount function.
These lines can be used to reproduce the issue (using /dev/sdm):
SCRATCH_DEV=/dev/sdm
SCRATCH_MNT=/mnt
echo 0 25165824 linear $SCRATCH_DEV 0 | dmsetup create foo
ls -alLF /dev/mapper/foo
mkfs.btrfs /dev/mapper/foo
mount /dev/mapper/foo $SCRATCH_MNT
echo bar > $SCRATCH_MNT/foo
sync
echo 0 25165824 error | dmsetup reload foo
dmsetup resume foo
ls -alF $SCRATCH_MNT
touch $SCRATCH_MNT/1
ls -alF $SCRATCH_MNT
sleep 35
echo 0 25165824 linear $SCRATCH_DEV 0 | dmsetup reload foo
dmsetup resume foo
sleep 1
umount $SCRATCH_MNT
btrfsck /dev/mapper/foo
dmsetup remove foo
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de> Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Josef Bacik [Tue, 7 Aug 2012 20:00:32 +0000 (16:00 -0400)]
Btrfs: allow delayed refs to be merged
Daniel Blueman reported a bug with fio+balance on a ramdisk setup.
Basically what happens is the balance relocates a tree block which will drop
the implicit refs for all of its children and adds a full backref. Once the
block is relocated we have to add the implicit refs back, so when we cow the
block again we add the implicit refs for its children back. The problem
comes when the original drop ref doesn't get run before we add the implicit
refs back. The delayed ref stuff will specifically prefer ADD operations
over DROP to keep us from freeing up an extent that will have references to
it, so we try to add the implicit ref before it is actually removed and we
panic. This worked fine before because the add would have just canceled the
drop out and we would have been fine. But the backref walking work needs to
be able to freeze the delayed ref stuff in time so we have this ever
increasing sequence number that gets attached to all new delayed ref updates
which makes us not merge refs and we run into this issue.
So to fix this we need to merge delayed refs. So everytime we run a
clustered ref we need to try and merge all of its delayed refs. The backref
walking stuff locks the delayed ref head before processing, so if we have it
locked we are safe to merge any refs inside of the sequence number. If
there is no sequence number we can merge all refs. Doing this not only
fixes our bug but keeps the delayed ref code from adding and removing
useless refs and batching together multiple refs into one search instead of
one search per delayed ref, which will really help our commit times. I ran
this with Daniels test and 276 and I haven't seen any problems. Thanks,
Reported-by: Daniel J Blueman <daniel@quora.org> Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Josef Bacik [Wed, 8 Aug 2012 16:12:59 +0000 (10:12 -0600)]
Btrfs: fix enospc problems when deleting a subvol
Subvol delete is a special kind of awful where we use the global reserve to
cover the ENOSPC requirements. The problem is once we're done removing
everything we do a btrfs_update_inode(), which by default will try to do the
delayed update stuff which will use it's own reserve. There will be no
space in this reserve and we'll return ENOSPC. So instead use
btrfs_update_inode_fallback() which will just fallback to updating the inode
item in the case of enospc. This is fine because the global reserve covers
the space requirements for this. With this patch I can now delete a subvol
on a problem image Dave Sterba sent me. Thanks,
Reported-by: David Sterba <dave@jikos.cz> Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
I added a window where the delayed_ref's head->ref_mod code can diverge
from the sum of the remaining refs, because we release the head->mutex
in the middle. This leads to btrfs_lookup_extent_info returning wrong
numbers. This patch fixes this by adjusting the head's ref_mod with each
delayed ref we run.
Signed-off-by: Arne Jansen <sensille@gmx.net> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Chris Mason [Tue, 7 Aug 2012 19:34:49 +0000 (15:34 -0400)]
Btrfs: don't run __tree_mod_log_free_eb on leaves
When we split a leaf, we may end up inserting a new root on top of that
leaf. The reflog code was incorrectly assuming the old root was always
a node. This makes sure we skip over leaves.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Josef Bacik [Mon, 6 Aug 2012 19:46:38 +0000 (13:46 -0600)]
Btrfs: increase the size of the free space cache
Arne was complaining about the space cache having mismatching generation
numbers when debugging a deadlock. This is because we can run out of space
in our preallocated range for our space cache if you have a pretty
fragmented amount of space in your pinned space. So just increase the
amount of space we preallocate for space cache so we can be sure to have
enough space. This will only really affect data ranges since their the only
chunks that end up larger than 256MB. Thanks,
Signed-off-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Josef Bacik [Wed, 1 Aug 2012 19:36:24 +0000 (15:36 -0400)]
Btrfs: barrier before waitqueue_active
We need a barrir before calling waitqueue_active otherwise we will miss
wakeups. So in places that do atomic_dec(); then atomic_read() use
atomic_dec_return() which imply a memory barrier (see memory-barriers.txt)
and then add an explicit memory barrier everywhere else that need them.
Thanks,
Arne Jansen [Mon, 6 Aug 2012 20:18:51 +0000 (14:18 -0600)]
Btrfs: fix deadlock in wait_for_more_refs
Commit a168650c introduced a waiting mechanism to prevent busy waiting in
btrfs_run_delayed_refs. This can deadlock with btrfs_run_ordered_operations,
where a tree_mod_seq is held while waiting for the io to complete, while
the end_io calls btrfs_run_delayed_refs.
This whole mechanism is unnecessary. If not enough runnable refs are
available to satisfy count, just return as count is more like a guideline
than a strict requirement.
In case we have to run all refs, commit transaction makes sure that no
other threads are working in the transaction anymore, so we just assert
here that no refs are blocked.
Signed-off-by: Arne Jansen <sensille@gmx.net> Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Josef Bacik [Fri, 3 Aug 2012 20:49:19 +0000 (16:49 -0400)]
Btrfs: don't allocate a seperate csums array for direct reads
We've been allocating a big array for csums instead of storing them in the
io_tree like we do for buffered reads because previously we were locking the
entire range, so we didn't have an extent state for each sector of the
range. But now that we do the range locking as we map the buffers we can
limit the mapping lenght to sectorsize and use the private part of the
io_tree for our csums. This allows us to avoid an extra memory allocation
for direct reads which could incur latency. Thanks,
Josef Bacik [Thu, 2 Aug 2012 14:23:59 +0000 (10:23 -0400)]
Btrfs: do not strdup non existent strings
When we close devices we add back empty devices for some reason that escapes
me. In the case of a missing dev we don't allocate an rcu_string for it's
name, so check to see if the device has a name and if it doesn't don't
bother strdup()'ing it. Thanks,
the box will panic trying to deref the name for the missing dev since it is
the lower numbered devid. So fix show_devname to not use missing devices.
Thanks,
Stefan Behrens [Wed, 1 Aug 2012 10:28:01 +0000 (04:28 -0600)]
Btrfs: fix that error value is changed by mistake
In iterate_inodes_from_logical() the error result from
extent_from_logical() is patched by mistake. Typically ENOENT is
patched to EINVAL because (-ENOENT & BTRFS_EXTENT_FLAG_TREE_BLOCK)
evaluates to true.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
This is because we would not return EIOCBQUEUED for short AIO reads, instead
we'd wait for the DIO to complete and then return the amount of data we
transferred, which would allow our stuff to unlock the remaning amount. But
with this change this no longer happens, so if we have a short AIO read (for
example if we try to read past EOF), we could leave the section from EOF to
the end of where we tried to read locked. Fixing this is tricky since there
is no clear way to know exactly how much data DIO truly submitted for IO, so
to make this less hard on ourselves and less combersome we need to lock the
extents as we try to map them, and then we unlock any areas we didn't
actually map. This makes us completely safe from deadlocks and reliance on
a particular behavior of the DIO code. This also lays the groundwork for
allowing us to use the normal csum storage method for reads which means we
can remove an allocation. Thanks,
projects / karo-tx-linux.git / log