28 Jul, 2016

1 commit

  • Pull LED updates from Jacek Anaszewski:
    "New LED class driver:
    - LED driver for TI LP3952 6-Channel Color LED

    LED core improvements:
    - Only descend into leds directory when CONFIG_NEW_LEDS is set
    - Add no-op gpio_led_register_device when LED subsystem is disabled
    - MAINTAINERS: Add file patterns for led device tree bindings

    LED Trigger core improvements:
    - return error if invalid trigger name is provided via sysfs

    LED class drivers improvements
    - is31fl32xx: define complete i2c_device_id table
    - is31fl32xx: fix typo in id and match table names
    - leds-gpio: Set of_node for created LED devices
    - pca9532: Add device tree support

    Conversion of IDE trigger to common disk trigger:
    - leds: convert IDE trigger to common disk trigger
    - leds: documentation: 'ide-disk' to 'disk-activity'
    - unicore32: use the new LED disk activity trigger
    - parisc: use the new LED disk activity trigger
    - mips: use the new LED disk activity trigger
    - arm: use the new LED disk activity trigger
    - powerpc: use the new LED disk activity trigger"

    * tag 'leds_for_4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds:
    leds: is31fl32xx: define complete i2c_device_id table
    leds: is31fl32xx: fix typo in id and match table names
    leds: LED driver for TI LP3952 6-Channel Color LED
    leds: leds-gpio: Set of_node for created LED devices
    leds: triggers: return error if invalid trigger name is provided via sysfs
    leds: Only descend into leds directory when CONFIG_NEW_LEDS is set
    leds: Add no-op gpio_led_register_device when LED subsystem is disabled
    unicore32: use the new LED disk activity trigger
    parisc: use the new LED disk activity trigger
    mips: use the new LED disk activity trigger
    arm: use the new LED disk activity trigger
    powerpc: use the new LED disk activity trigger
    leds: documentation: 'ide-disk' to 'disk-activity'
    leds: convert IDE trigger to common disk trigger
    leds: pca9532: Add device tree support
    MAINTAINERS: Add file patterns for led device tree bindings

    Linus Torvalds
     

27 Jul, 2016

2 commits

  • Pull core block updates from Jens Axboe:

    - the big change is the cleanup from Mike Christie, cleaning up our
    uses of command types and modified flags. This is what will throw
    some merge conflicts

    - regression fix for the above for btrfs, from Vincent

    - following up to the above, better packing of struct request from
    Christoph

    - a 2038 fix for blktrace from Arnd

    - a few trivial/spelling fixes from Bart Van Assche

    - a front merge check fix from Damien, which could cause issues on
    SMR drives

    - Atari partition fix from Gabriel

    - convert cfq to highres timers, since jiffies isn't granular enough
    for some devices these days. From Jan and Jeff

    - CFQ priority boost fix idle classes, from me

    - cleanup series from Ming, improving our bio/bvec iteration

    - a direct issue fix for blk-mq from Omar

    - fix for plug merging not involving the IO scheduler, like we do for
    other types of merges. From Tahsin

    - expose DAX type internally and through sysfs. From Toshi and Yigal

    * 'for-4.8/core' of git://git.kernel.dk/linux-block: (76 commits)
    block: Fix front merge check
    block: do not merge requests without consulting with io scheduler
    block: Fix spelling in a source code comment
    block: expose QUEUE_FLAG_DAX in sysfs
    block: add QUEUE_FLAG_DAX for devices to advertise their DAX support
    Btrfs: fix comparison in __btrfs_map_block()
    block: atari: Return early for unsupported sector size
    Doc: block: Fix a typo in queue-sysfs.txt
    cfq-iosched: Charge at least 1 jiffie instead of 1 ns
    cfq-iosched: Fix regression in bonnie++ rewrite performance
    cfq-iosched: Convert slice_resid from u64 to s64
    block: Convert fifo_time from ulong to u64
    blktrace: avoid using timespec
    block/blk-cgroup.c: Declare local symbols static
    block/bio-integrity.c: Add #include "blk.h"
    block/partition-generic.c: Remove a set-but-not-used variable
    block: bio: kill BIO_MAX_SIZE
    cfq-iosched: temporarily boost queue priority for idle classes
    block: drbd: avoid to use BIO_MAX_SIZE
    block: bio: remove BIO_MAX_SECTORS
    ...

    Linus Torvalds
     
  • Pull libata updates from Tejun Heo:
    "libata saw quite a bit of activities in this cycle:

    - SMR drive support still being worked on

    - bug fixes and improvements to misc SCSI command emulation

    - some low level driver updates"

    * 'for-4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata: (39 commits)
    libata-scsi: better style in ata_msense_*()
    AHCI: Clear GHC.IS to prevent unexpectly asserting INTx
    ata: sata_dwc_460ex: remove redundant dev_err call
    ata: define ATA_PROT_* in terms of ATA_PROT_FLAG_*
    libata: remove ATA_PROT_FLAG_DATA
    libata: remove ata_is_nodata
    ata: make lba_{28,48}_ok() use ATA_MAX_SECTORS{,_LBA48}
    libata-scsi: minor cleanup for ata_scsi_zbc_out_xlat
    libata-scsi: Fix ZBC management out command translation
    libata-scsi: Fix translation of REPORT ZONES command
    ata: Handle ATA NCQ NO-DATA commands correctly
    libata-eh: decode all taskfile protocols
    ata: fixup ATA_PROT_NODATA
    libsas: use ata_is_ncq() and ata_has_dma() accessors
    libata: use ata_is_ncq() accessors
    libata: return boolean values from ata_is_*
    libata-scsi: avoid repeated calculation of number of TRIM ranges
    libata-scsi: reject WRITE SAME (16) with n_block that exceeds limit
    libata-scsi: rename ata_msense_ctl_mode() to ata_msense_control()
    libata-scsi: fix D_SENSE bit relection in control mode page
    ...

    Linus Torvalds
     

23 Jul, 2016

1 commit


20 Jul, 2016

3 commits

  • `changeable` is the "version" of mode page requested by the user.
    It will be less confusing/misleading if we do not check it
    "together" with the setting bits of the drive.

    Not to mention that we currently have ata_mselect_*() implemented
    in a way that each of them will serve exclusively a particular bit
    on each page. The old style will hence make the condition look even
    more unnecessarily arcane if the ata_msense_*() is reflecting more
    than one bit.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • Due to PCI subsystem behaviour, unloading AHCI driver will disable
    MSI and enable INTx. When HBA supports MSIx or Multiple MSI, Driver's
    irq handler doesn't clear GHC.IS register. It works well when reading or
    writing data and GHC.IS is always non-zero. But when unloading driver
    (or any other operation which causes disable MSIx and enable INTx), PCI
    subsystem uses config write(Rx04.bit10) to enable INTx. Because
    GHC.IS is non-zero, HBA will falsely assume some port needs interrupt
    service. Then it asserts INTx. To make things worse, when AHCI controller
    shares the same interrupt pin with other PCI device, that PCI device's ISR
    will be called and nobody de-asserts previous INTx.
    This patch clears GHC.IS in ahci_port_stop() even when using MSIx or
    MMSI to prevent this case. It ensures GHC.IS is zero before PCI subsystem
    enables INTx.

    tj: Minor updates to the comment.

    Signed-off-by: Raymond Pang
    Signed-off-by: Tejun Heo

    Pang Raymond
     
  • There is a error message within devm_ioremap_resource
    already, so remove the dev_err call to avoid redundant
    error message.

    Signed-off-by: Wei Yongjun
    Signed-off-by: Tejun Heo

    Wei Yongjun
     

19 Jul, 2016

3 commits


15 Jul, 2016

4 commits

  • The reset_all variable name is misleading as this bit is also applicable to
    open, close, and finish actions. So rename that variable to "all" and remove
    the unnecessary mask operation that's already done earlier.

    Signed-off-by: Damien Le Moal
    [hch: split from the previous patch]
    Signed-off-by: Christoph Hellwig
    Signed-off-by: Tejun Heo

    Damien Le Moal
     
  • The subcommand for NCQ NON-DATA must be specified in the feature
    (low byte), not the high-order count byte. Also make sure to properly
    cast the all bit to a u16 before shiting it by 8 to avoid undefined
    behavior.

    Signed-off-by: Damien Le Moal
    [hch: split the original patch into two, updated changelog]
    Signed-off-by: Christoph Hellwig
    Signed-off-by: Tejun Heo

    Damien Le Moal
     
  • Include reporting options when translating REPORT ZONES commmand to
    ATA NCQ, and make sure we only look at the actually specified bits
    in the CDB for the options.

    Signed-off-by: Damien Le Moal
    [hch: update patch description]
    Signed-off-by: Christoph Hellwig
    Signed-off-by: Tejun Heo

    Damien Le Moal
     
  • Add a new taskfile protocol ATA_PROT_NCQ_NODATA to handle
    ATA NCQ NO-DATA commands correctly.
    And fixup ata_scsi_zbc_out_xlat() to use it.

    Signed-off-by: Hannes Reinecke
    Signed-off-by: Christoph Hellwig
    Signed-off-by: Tejun Heo

    Hannes Reinecke
     

14 Jul, 2016

3 commits


13 Jul, 2016

4 commits

  • Currently libata statically allows only 1-block (512-byte) payload
    for each TRIM command. Each payload can carry 64 TRIM ranges since
    each range requires 8 bytes.

    It is silly to keep doing the calculation (512 / 8) in different
    places. Hence, define the new ATA_MAX_TRIM_RNUM for the result.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • Currently if a WRITE SAME (16) command is issued to the SATL with
    "number of blocks" that is larger than the "Maximum write same length"
    (which is the maximum number of blocks per TRIM command allowed in
    libata, currently 65535 * 512 / 8 blocks), the SATL will accept the
    command and translate it to a TRIM command with the upper limit.

    However, according to SBC (as of sbc4r11.pdf), the "device server"
    should terminate the command with "Invalid field in CDB" in that case.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • To make it consistent with the recently added ata_mselect_control().
    We probably shouldn't have the word "mode" in its name anyway, since
    that's not the case for other ata_msense_*() / ata_mselect_*() either.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • The bit should always be set to 1 when the requested version of
    page is "changeable" because we've made it so in ata_mselect_control().
    Also, it should always be set to 1 if ATA_DFLAG_D_SENSE is set (when
    the requested version of page is "current" or "default").

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     

12 Jul, 2016

5 commits

  • According to the HPT366 data sheet, PCI config space dword 0x40-0x43
    bits 11:8 specify the primary drive cmd_high_time, however,
    currently just 3 bits of the 4 are being used because the mask
    is 0x700 and not 0x0f00. Fix the mask, allowing for the 40MHz clock
    to be detected.

    Signed-off-by: Colin Ian King
    Signed-off-by: Tejun Heo

    Colin Ian King
     
  • The comment suggests we should be having an SPC-3 version descriptor
    but the 0260h is the code for "SPC-2 (no version claimed)". Correct
    it to 0300h so that it has the "SPC-3 (no version claimed)" descriptor.

    Note that we are claiming SPC-3 version compatibility in the VERSION
    field of the standard INQUIRY data. Therefore, I assume the typo was
    on the code but not on the comment.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • Avoid performance bottleneck when being SCSI pass-through'd to
    virtual machines with other OSes (e.g. Windows) via virtio-scsi
    and scsi-block in qemu.

    Ref.: https://github.com/YanVugenfirer/kvm-guest-drivers-windows/issues/63

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • Without this fix, the DRA bit of the caching mode page would not
    be updated when the read look-ahead feature is toggled (e.g. with
    `smartctl --set`), but will only be until, for example, the write
    cache feature is touched.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     
  • Commit 856c46639309 ("libata: support device-managed ZAC devices")
    had the line that "bumps" the VERSION field in standard INQUIRY data
    removed. Add it back and claim SPC-5 version compatibility, which
    matches with the current version descriptor "SPC-5 (no version claimed)"
    that is used for ZAC devices.

    Signed-off-by: Tom Yan
    Signed-off-by: Tejun Heo

    Tom Yan
     

07 Jul, 2016

2 commits


29 Jun, 2016

2 commits


20 Jun, 2016

1 commit

  • This patch converts the IDE specific LED trigger to a generic disk
    activity LED trigger. The libata core is now a trigger source just
    like before the IDE disk driver. It's merely a replacement of the
    string ide by disk.

    The patch is taken from http://dev.gentoo.org/~josejx/ata.patch and is
    widely used by any ibook/powerbook owners with great satisfaction.
    Likewise, it is very often used successfully on different ARM platforms.

    Unlike the original patch, the existing 'ide-disk' trigger is still
    available for backward compatibility. That reduce the amount of patches
    in affected device trees out of the mainline kernel. For further
    development, the new name 'disk-activity' should be used.

    Cc: Joseph Jezak
    Cc: Jörg Sommer
    Cc: Richard Purdie
    Signed-off-by: Stephan Linz
    Signed-off-by: Jacek Anaszewski

    Stephan Linz
     

19 Jun, 2016

1 commit


18 Jun, 2016

2 commits

  • Fix the signed issue in mv_write_cached_reg() where the laddr
    is assigned from a (long)addr instead of (unsigned long)addr.

    Fixes the following warnings:

    drivers/ata/sata_mv.c:989:26: warning: cast removes address space of expression
    drivers/ata/sata_mv.c:989:26: warning: cast removes address space of expression
    drivers/ata/sata_mv.c:989:26: warning: cast removes address space of expression
    drivers/ata/sata_mv.c:989:26: warning: cast removes address space of expression

    Signed-off-by: Ben Dooks
    Signed-off-by: Tejun Heo

    Ben Dooks
     
  • The sysfs file for the libata error handling has multiple issues
    in the way it prints time stamps:

    * it prints a 9-digit nanosecond value using a %06lu format string,
    which drops some leading zeroes
    * it converts a 64-bit jiffes value to a timespec using
    jiffies_to_timespec(), which takes a 'long' argument, so the
    result is wrong after a jiffies overflow (49 days).
    * we try to avoid using timespec because that generally overflows
    in 2038, although this particular usage is ok.

    This replaces the jiffies_to_timespec call with an open-coded
    implementation that gets it right.

    Signed-off-by: Arnd Bergmann
    Signed-off-by: Tejun Heo

    Arnd Bergmann
     

17 Jun, 2016

2 commits


16 Jun, 2016

1 commit


09 Jun, 2016

1 commit

  • sas_ata_strategy_handler() adds the works of the ata error handler to
    system_unbound_wq. This workqueue asynchronously runs work items, so the
    ata error handler will be performed concurrently on different CPUs. In
    this case, ->host_failed will be decreased simultaneously in
    scsi_eh_finish_cmd() on different CPUs, and become abnormal.

    It will lead to permanently inequality between ->host_failed and
    ->host_busy, and scsi error handler thread won't start running. IO
    errors after that won't be handled.

    Since all scmds must have been handled in the strategy handler, just
    remove the decrement in scsi_eh_finish_cmd() and zero ->host_busy after
    the strategy handler to fix this race.

    Fixes: 50824d6c5657 ("[SCSI] libsas: async ata-eh")
    Cc: stable@vger.kernel.org
    Signed-off-by: Wei Fang
    Reviewed-by: James Bottomley
    Signed-off-by: Martin K. Petersen

    Wei Fang
     

08 Jun, 2016

1 commit

  • We currently set REQ_WRITE/WRITE for all non READ IOs
    like discard, flush, writesame, etc. In the next patches where we
    no longer set up the op as a bitmap, we will not be able to
    detect a operation direction like writesame by testing if REQ_WRITE is
    set.

    This patch converts the drivers and cgroup to use the
    op_is_write helper. This should just cover the simple
    cases. I did dm, md and bcache in their own patches
    because they were more involved.

    Signed-off-by: Mike Christie
    Reviewed-by: Hannes Reinecke
    Signed-off-by: Jens Axboe

    Mike Christie
     

28 May, 2016

1 commit

  • Most users of IS_ERR_VALUE() in the kernel are wrong, as they
    pass an 'int' into a function that takes an 'unsigned long'
    argument. This happens to work because the type is sign-extended
    on 64-bit architectures before it gets converted into an
    unsigned type.

    However, anything that passes an 'unsigned short' or 'unsigned int'
    argument into IS_ERR_VALUE() is guaranteed to be broken, as are
    8-bit integers and types that are wider than 'unsigned long'.

    Andrzej Hajda has already fixed a lot of the worst abusers that
    were causing actual bugs, but it would be nice to prevent any
    users that are not passing 'unsigned long' arguments.

    This patch changes all users of IS_ERR_VALUE() that I could find
    on 32-bit ARM randconfig builds and x86 allmodconfig. For the
    moment, this doesn't change the definition of IS_ERR_VALUE()
    because there are probably still architecture specific users
    elsewhere.

    Almost all the warnings I got are for files that are better off
    using 'if (err)' or 'if (err < 0)'.
    The only legitimate user I could find that we get a warning for
    is the (32-bit only) freescale fman driver, so I did not remove
    the IS_ERR_VALUE() there but changed the type to 'unsigned long'.
    For 9pfs, I just worked around one user whose calling conventions
    are so obscure that I did not dare change the behavior.

    I was using this definition for testing:

    #define IS_ERR_VALUE(x) ((unsigned long*)NULL == (typeof (x)*)NULL && \
    unlikely((unsigned long long)(x) >= (unsigned long long)(typeof(x))-MAX_ERRNO))

    which ends up making all 16-bit or wider types work correctly with
    the most plausible interpretation of what IS_ERR_VALUE() was supposed
    to return according to its users, but also causes a compile-time
    warning for any users that do not pass an 'unsigned long' argument.

    I suggested this approach earlier this year, but back then we ended
    up deciding to just fix the users that are obviously broken. After
    the initial warning that caused me to get involved in the discussion
    (fs/gfs2/dir.c) showed up again in the mainline kernel, Linus
    asked me to send the whole thing again.

    [ Updated the 9p parts as per Al Viro - Linus ]

    Signed-off-by: Arnd Bergmann
    Cc: Andrzej Hajda
    Cc: Andrew Morton
    Link: https://lkml.org/lkml/2016/1/7/363
    Link: https://lkml.org/lkml/2016/5/27/486
    Acked-by: Srinivas Kandagatla # For nvmem part
    Signed-off-by: Linus Torvalds

    Arnd Bergmann