aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* NFS: Fix nfs4_verifier memory alignmentChuck Lever2012-03-06
| | | | | | | | | | | | | | | | | | | | | | | Clean up due to code review. The nfs4_verifier's data field is not guaranteed to be u32-aligned. Casting an array of chars to a u32 * is considered generally hazardous. Fix this by using a __be32 array to generate a verifier's contents, and then byte-copy the contents into the verifier field. The contents of a verifier, for all intents and purposes, are opaque bytes. Only local code that generates a verifier need know the actual content and format. Everyone else compares the full byte array for exact equality. Also, sizeof(nfs4_verifer) is the size of the in-core verifier data structure, but NFS4_VERIFIER_SIZE is the number of octets in an XDR'd verifier. The two are not interchangeable, even if they happen to have the same value. Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Add a encode op helperTrond Myklebust2012-03-06
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Add a helper for encoding NFSv4 sequence idsTrond Myklebust2012-03-06
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Minor clean ups for encode_string()Trond Myklebust2012-03-06
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Simplify the struct nfs4_stateidTrond Myklebust2012-03-06
| | | | | | | | | Replace the union with the common struct stateid4 as defined in both RFC3530 and RFC5661. This makes it easier to access the sequence id, which will again make implementing support for parallel OPEN calls easier. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Add helpers for basic copying of stateidsTrond Myklebust2012-03-06
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Rename nfs4_copy_stateid()Trond Myklebust2012-03-06
| | | | | | | It is really a function for selecting the correct stateid to use in a read or write situation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Add a helper for encoding stateidsTrond Myklebust2012-03-06
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Add a helper for encoding opaque dataTrond Myklebust2012-03-06
| | | | Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Rename encode_stateid() to encode_open_stateid()Trond Myklebust2012-03-06
| | | | | | | The current version of encode_stateid really only applies to open stateids. You can't use it for locks, delegations or layouts. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4: Further clean-ups of delegation stateid validationTrond Myklebust2012-03-06
| | | | | | | | | Change the name to reflect what we're really doing: testing two stateids for whether or not they match according the the rules in RFC3530 and RFC5661. Move the code from callback_proc.c to nfs4proc.c Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFSv4.1: Fix matching of the stateids when returning a delegationTrond Myklebust2012-03-06
| | | | | | | | | | | | nfs41_validate_delegation_stateid is broken if we supply a stateid with a non-zero sequence id. Instead of trying to match the sequence id, the function assumes that we always want to error. While this is true for a delegation callback, it is not true in general. Also fix a typo in nfs4_callback_recall. Reported-by: Andy Adamson <andros@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Properly handle the case where the delegation is revokedTrond Myklebust2012-03-06
| | | | | | | | | | | | | | | | If we know that the delegation stateid is bad or revoked, we need to remove that delegation as soon as possible, and then mark all the stateids that relied on that delegation for recovery. We cannot use the delegation as part of the recovery process. Also note that NFSv4.1 uses a different error code (NFS4ERR_DELEG_REVOKED) to indicate that the delegation was revoked. Finally, ensure that setlk() and setattr() can both recover safely from a revoked delegation. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: stable@vger.kernel.org
* NFS: Fix a typo in _nfs_display_fhandleTrond Myklebust2012-03-06
| | | | | | | | The check for 'fh == NULL' needs to come _before_ we dereference fh. Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Fix a compile issue when !CONFIG_NFS_V4_1Trond Myklebust2012-03-05
| | | | | | | | The attempt to display the implementation ID needs to be conditional on whether or not CONFIG_NFS_V4_1 is defined Reported-by: Bryan Schumaker <Bryan.Schumaker@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* NFS: Undo changes to idmap.hBryan Schumaker2012-03-05
| | | | | | | | When compiled without NFS v4 configured these function won't be defined and the compiler will yell. Signed-off-by: Bryan Schumaker <bjschuma@netapp.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* Merge commit 'nfs-for-3.3-4' into nfs-for-nextTrond Myklebust2012-03-03
|\ | | | | | | | | | | | | | | Conflicts: fs/nfs/nfs4proc.c Back-merge of the upstream kernel in order to fix a conflict with the slotid type conversion and implementation id patches...
| * NFSv4: fix server_scope memory leakWeston Andros Adamson2012-02-17
| | | | | | | | | | | | | | | | | | server_scope would never be freed if nfs4_check_cl_exchange_flags() returned non-zero Signed-off-by: Weston Andros Adamson <dros@netapp.com> Cc: stable@vger.kernel.org Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
| * NFSv4.1: Fix a NFSv4.1 session initialisation regressionTrond Myklebust2012-02-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit aacd553 (NFSv4.1: cleanup init and reset of session slot tables) introduces a regression in the session initialisation code. New tables now find their sequence ids initialised to 0, rather than the mandated value of 1 (see RFC5661). Fix the problem by merging nfs4_reset_slot_table() and nfs4_init_slot_table(). Since the tbl->max_slots is initialised to 0, the test in nfs4_reset_slot_table for max_reqs != tbl->max_slots will automatically pass for an empty table. Reported-by: Vitaliy Gusev <gusev.vitaliy@nexenta.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
| * NFSv4: Ensure we throw out bad delegation stateids on NFS4ERR_BAD_STATEIDTrond Myklebust2012-02-09
| | | | | | | | | | | | | | | | To ensure that we don't just reuse the bad delegation when we attempt to recover the nfs4_state that received the bad stateid error. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: stable@vger.kernel.org
| * NFSv4: Fix an Oops in the NFSv4 getacl codeTrond Myklebust2012-02-03
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit bf118a342f10dafe44b14451a1392c3254629a1f (NFSv4: include bitmap in nfsv4 get acl data) introduces the 'acl_scratch' page for the case where we may need to decode multi-page data. However it fails to take into account the fact that the variable may be NULL (for the case where we're not doing multi-page decode), and it also attaches it to the encoding xdr_stream rather than the decoding one. The immediate result is an Oops in nfs4_xdr_enc_getacl due to the call to page_address() with a NULL page pointer. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: Andy Adamson <andros@netapp.com> Cc: stable@vger.kernel.org
| * Merge tag 'kmemleak-fixes' of ↵Linus Torvalds2012-02-03
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux Trivial kmemleak bug-fixes: - Early logging doesn't stop when kmemleak is off by default. - Zero-size scanning areas should be ignored (currently it prints a warning). * tag 'kmemleak-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux: kmemleak: Disable early logging when kmemleak is off by default kmemleak: Only scan non-zero-size areas
| | * kmemleak: Disable early logging when kmemleak is off by defaultCatalin Marinas2012-01-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit b6693005 (kmemleak: When the early log buffer is exceeded, report the actual number) deferred the disabling of the early logging to kmemleak_init(). However, when CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, the early logging was no longer disabled causing __init kmemleak functions to be called even after the kernel freed the init memory. This patch disables the early logging during kmemleak_init() if kmemleak is left disabled. Reported-by: Dirk Gouders <gouders@et.bocholt.fh-gelsenkirchen.de> Tested-by: Dirk Gouders <gouders@et.bocholt.fh-gelsenkirchen.de> Tested-by: Josh Boyer <jwboyer@gmail.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
| | * kmemleak: Only scan non-zero-size areasTiejun Chen2012-01-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | Kmemleak should only track valid scan areas with a non-zero size. Otherwise, such area may reside just at the end of an object and kmemleak would report "Adding scan area to unknown object". Signed-off-by: Tiejun Chen <tiejun.chen@windriver.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
| * | Merge tag 'sound-fixes' of ↵Linus Torvalds2012-02-03
| |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound sound fixes for 3.3-rc3 Most of commits are either regression fixes for varioud HD-audio codecs or small ASoC fixes. Also a trivial build fix is included. * tag 'sound-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: ALSA: hda - Disable dynamic-power control for VIA as default ALSA: hda - Allow analog low-current mode when dynamic power-control is on ALSA: hda - Fix the logic to detect VIA analog low-current mode ALSA: hda - Check power-state before changing in patch_via.c ALSA: HDA: Fix duplicated output to more than one codec ALSA: hda - Fix calling cs_automic twice for Cirrus codecs. ALSA: HDA: Remove quirk for Toshiba Qosmio G50 ALSA: HDA: Fix jack creation for codecs with front and rear Line In ALSA: hda - Apply 0x0f-VREF fix to all ASUS laptops with ALC861/660 ASoC: neo1973_wm8753: remove references to the neo1973-gta01 machine ALSA: Add #ifdef CONFIG_PCI guard for snd_pci_quirk_* functions ASoC: wm_hubs: fix wrong bits for LINEOUT2 N/P mixer ALSA: HDA: Remove quirk for Asus N53Jq ASoC: wm_hubs: Enable line out VMID buffer for single ended line outputs ASoC: wm5100: Mark register cache as dirty when regulators are disabled ASoC: wm8962: Mark register cache as dirty when regulators are disabled ASoC: wm8996: Mark register cache as dirty when regulators are disabled ASoC: wm5100: Fix microphone configuration ASoC: wm5100: Make sure we switch to button reporting mode
| | * | ALSA: hda - Disable dynamic-power control for VIA as defaultTakashi Iwai2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since the dynamic pin power-control and the analog low-current mode may lead to pop-noise, it's safer to set it off as default. Bugzilla: https://bugzilla.novell.com/show_bug.cgi?id=741128 Cc: <stable@kernel.org> [v3.1+] Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | ALSA: hda - Allow analog low-current mode when dynamic power-control is onTakashi Iwai2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | VIA codecs have several different power-saving features, and one of them is the analog low-current mode. But it turned out that the ALC mode causes pop-noises at each on/off time on some machines. As a quick workaround, disable the ALC when another power-saving feature, the dynamic pin power-control, is turned off, too, since the dynamic power-control is already exposed as a mixer enum element so that user can turn it on/off freely. Bugzilla: https://bugzilla.novell.com/show_bug.cgi?id=741128 Cc: <stable@kernel.org> [v3.1+] Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | ALSA: hda - Fix the logic to detect VIA analog low-current modeTakashi Iwai2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The analog low-current mode must be enabled when the no stream is running but the current detection checks it in a wrong way. Bugzilla: https://bugzilla.novell.com/show_bug.cgi?id=741128 Cc: <stable@kernel.org> [v3.1+] Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | ALSA: hda - Check power-state before changing in patch_via.cTakashi Iwai2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of always writing AC_VERB_SET_POWER_STATE, check the current power-state and don't write again if the value is already set. This may reduce the click noise upon the dynamic power-state change (e.g. in analog-input mixer). Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | ALSA: HDA: Fix duplicated output to more than one codecDavid Henningsson2012-02-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This typo caused the wrong codec's nid to be checked for wcaps type. As a result, sometimes speakers would duplicate the output sent to HDMI output. Cc: stable@kernel.org BugLink: https://bugs.launchpad.net/bugs/924320 Signed-off-by: David Henningsson <david.henningsson@canonical.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | ALSA: hda - Fix calling cs_automic twice for Cirrus codecs.Dylan Reid2012-02-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If cs_automic is called twice (like it is during init) while the mic is present, it will over-write the last_input with the new one, causing it to switch back to the automic input when the mic is unplugged. This leaves the driver in a state (cur_input, last_input, and automix_idx the same) where the internal mic can not be selected until it is rebooted without the mic attached. Check that the mic hasn't already been switched to before setting last_input. Signed-off-by: Dylan Reid <dgreid@chromium.org> Cc: <stable@kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | Merge branch 'fix/asoc' into for-linusTakashi Iwai2012-01-31
| | |\ \
| | | * \ Merge tag 'asoc-3.3' of ↵Takashi Iwai2012-01-30
| | | |\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into fix/asoc A bunch of small driver specific fixes - nothing terribly exciting here, all of this will only affect people using particular devices and then usually only in some use cases.
| | | | * | ASoC: neo1973_wm8753: remove references to the neo1973-gta01 machineDenis 'GNUtoo' Carikli2012-01-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Openmoko GTA01 machine has been removed from the machine ID database, so we need to remove references to it as well. Without that fix we have: sound/soc/samsung/neo1973_wm8753.c: In function ‘neo1973_wm8753_init’: sound/soc/samsung/neo1973_wm8753.c:325:2: error: implicit declaration of function ‘machine_is_neo1973_gta01’ Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo@no-log.org> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| | | | * | ASoC: wm_hubs: fix wrong bits for LINEOUT2 N/P mixerUK KIM2012-01-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: UK KIM <w0806.kim@samsung.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Cc: stable@vger.kernel.org
| | | | * | ASoC: wm_hubs: Enable line out VMID buffer for single ended line outputsMark Brown2012-01-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For optimal performance the single ended line outputs require that the line output VMID buffer be enabled. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Cc: stable@vger.kernel.org
| | | | * | ASoC: wm5100: Mark register cache as dirty when regulators are disabledMark Brown2012-01-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Otherwise we won't resync later. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| | | | * | ASoC: wm8962: Mark register cache as dirty when regulators are disabledMark Brown2012-01-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Otherwise we won't resync later. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| | | | * | ASoC: wm8996: Mark register cache as dirty when regulators are disabledMark Brown2012-01-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Otherwise we won't resync later. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| | | | * | ASoC: wm5100: Fix microphone configurationMark Brown2012-01-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We need to write the configuration for each microphone to a different register. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Cc: stable@vger.kernel.org
| | | | * | ASoC: wm5100: Make sure we switch to button reporting modeMark Brown2012-01-25
| | | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we have identified an accessory make sure we've flagged that we've done so in order to make sure we always report buttons and don't continue to polarity flip. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
| | * | | ALSA: HDA: Remove quirk for Toshiba Qosmio G50David Henningsson2012-01-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The user reports that model=auto works better than current handling on a 3.2 based kernel (with jack detection patches backported). Since model=auto is what we prefer these days anyway, the quirk should be removed. Alsa-info for the relevant machine: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/923316/+attachment/2702812/+files/alsa-info.txt.Pbfno2x7bp BugLink: https://bugs.launchpad.net/bugs/923316 Signed-off-by: David Henningsson <david.henningsson@canonical.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | | ALSA: HDA: Fix jack creation for codecs with front and rear Line InDavid Henningsson2012-01-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If a codec has both a front and a rear Line In, two controls both named "Line Jack" will be created, which causes parsing to fail. While a long term solution might be to name the jacks differently, this extra check is consistent with what is already being done in many auto-parsers, and will also protect against other cases when two inputs have the same label. BugLink: https://bugs.launchpad.net/bugs/923409 Signed-off-by: David Henningsson <david.henningsson@canonical.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | | ALSA: hda - Apply 0x0f-VREF fix to all ASUS laptops with ALC861/660Takashi Iwai2012-01-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It turned out that other ASUS laptops require the similar fix to enable the VREF on the pin 0x0f for the secret output amp, not only ASUS A6Rp. Moreover, it's required even when the pin is being used as the output. Thus, writing a fixed value doesn't work always. This patch applies the VREF-fix for all ASUS laptops with ALC861/660 in a fixup function that checks the current value and turns on only the VREF value no matter whether input or output direction is set. The automute function is modified as well to keep the pin VREF upon muting/unmuting via pin-control; otherwise the pin VREF is reset at plugging/unplugging a jack. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=42588 Cc: <stable@kernel.org> [v3.2+] Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | | ALSA: Add #ifdef CONFIG_PCI guard for snd_pci_quirk_* functionsAxel Lin2012-01-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixes below build warning when CONFIG_PCI is not set. CC sound/sound_core.o In file included from sound/sound_core.c:15: include/sound/core.h:454: warning: 'struct pci_dev' declared inside parameter list include/sound/core.h:454: warning: its scope is only this definition or declaration, which is probably not what you want Signed-off-by: Axel Lin <axel.lin@gmail.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| | * | | ALSA: HDA: Remove quirk for Asus N53JqDavid Henningsson2012-01-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The user reports that he needs to add model=auto for audio to work properly. In fact, since node 0x15 is not even a pin node, the existing fixup is definitely wrong. Relevant information can be found in the buglink below. Cc: stable@kernel.org (3.2+) BugLink: https://bugs.launchpad.net/bugs/918254 Signed-off-by: David Henningsson <david.henningsson@canonical.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | | Merge branch 'for-linus' of ↵Linus Torvalds2012-02-02
| |\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: rbd: fix safety of rbd_put_client() rbd: fix a memory leak in rbd_get_client() ceph: create a new session lock to avoid lock inversion ceph: fix length validation in parse_reply_info() ceph: initialize client debugfs outside of monc->mutex ceph: change "ceph.layout" xattr to be "ceph.file.layout"
| | * | | | rbd: fix safety of rbd_put_client()Alex Elder2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The rbd_client structure uses a kref to arrange for cleaning up and freeing an instance when its last reference is dropped. The cleanup routine is rbd_client_release(), and one of the things it does is delete the rbd_client from rbd_client_list. It acquires node_lock to do so, but the way it is done is still not safe. The problem is that when attempting to reuse an existing rbd_client, the structure found might already be in the process of getting destroyed and cleaned up. Here's the scenario, with "CLIENT" representing an existing rbd_client that's involved in the race: Thread on CPU A | Thread on CPU B --------------- | --------------- rbd_put_client(CLIENT) | rbd_get_client() kref_put() | (acquires node_lock) kref->refcount becomes 0 | __rbd_client_find() returns CLIENT calls rbd_client_release() | kref_get(&CLIENT->kref); | (releases node_lock) (acquires node_lock) | deletes CLIENT from list | ...and starts using CLIENT... (releases node_lock) | and frees CLIENT | <-- but CLIENT gets freed here Fix this by having rbd_put_client() acquire node_lock. The result could still be improved, but at least it avoids this problem. Signed-off-by: Alex Elder <elder@dreamhost.com> Signed-off-by: Sage Weil <sage@newdream.net>
| | * | | | rbd: fix a memory leak in rbd_get_client()Alex Elder2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If an existing rbd client is found to be suitable for use in rbd_get_client(), the rbd_options structure is not being freed as it should. Fix that. Signed-off-by: Alex Elder <elder@dreamhost.com> Signed-off-by: Sage Weil <sage@newdream.net>
| | * | | | ceph: create a new session lock to avoid lock inversionAlex Elder2012-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Lockdep was reporting a possible circular lock dependency in dentry_lease_is_valid(). That function needs to sample the session's s_cap_gen and and s_cap_ttl fields coherently, but needs to do so while holding a dentry lock. The s_cap_lock field was being used to protect the two fields, but that can't be taken while holding a lock on a dentry within the session. In most cases, the s_cap_gen and s_cap_ttl fields only get operated on separately. But in three cases they need to be updated together. Implement a new lock to protect the spots updating both fields atomically is required. Signed-off-by: Alex Elder <elder@dreamhost.com> Reviewed-by: Sage Weil <sage@newdream.net>
generated by cgit v1.2.2 (git 2.25.0) at 2025-03-31 20:21:40 -0400