Merge tag 'ata-6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/libata...

Pull ata updates from Damien Le Moal:

 - Replace the ATA_DFLAG_ZAC device flag with the helper function
   ata_dev_is_zac() testing directly the device class and device zoned
   mode (me)

 - Some small cleanup of ata_scsi_offline_dev() code (me)

 - Improve the description of the link power management (LPM) policies
   in Kconfig and in the comments defining these. Together with this,
   clarify the description of the ahci driver mobile_lpm_policy module
   parameter (me)

 - Various code refactoring of libata LPM handling (ata_eh_set_lpm()
   renaming, introduce ata_dev_config_lpm(), LPM related quirk handling,
   and LPM related feature advertizing on device scan) (me)

 - Avoid unnecessary device reset when revalidating after an error when
   LPM is used (me)

 - Do not allow setting a port/link LPM policy if LPM is not supported,
   either because the controller does not support partial, slumber nor
   devsleep, or when the port is an external port with hotplug
   capability (me)

 - Make sure that device initiated power management (DIPM) is not
   enabled if the host (controller) lacks support for this feature (me)

 - Improve messages and debug messages related to LPM, in particular,
   reduce the number of messages signaling the lack of LPM support (me)

 - Cache in memory a device general purpose log directory to avoid
   having to access this log for every log page access. The intent here
   is to reduce the number of read log commands when scanning or
   revalidating a device (me)

 - Change ata_dev_cleanup_cdl_resources() to be a static function (me)

 - Rename and simplify the mode setting functions (me)

 - Introduce the helper function ata_port_eh_scheduled() to check if EH
   is pending or running for a port (me)

 - Improve ata_eh_set_pending() (return bool instead of int) (me)

 - Use sysfs_emit() instead of scnprintf() for libata-transport
   attributes (Jonathan)

 - Use the existing macro definiton of RDC vendor ID instead of
   hardcoding it in the pata_rdc driver (Andy)

 - Rework how EH is called for a port to avoid needing to pass along the
   prereset, softreset, hardreset and postreset operations. The driver
   API documentation for this is also updated (me)

* tag 'ata-6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux: (28 commits)
  Documentation: driver-api: Update libata error handler information
  ata: libata-eh: Simplify reset operation management
  ata: libata-eh: Remove ata_do_eh()
  ata: pata_rdc: Use registered definition for the RDC vendor
  ata: libata-eh: Make ata_eh_followup_srst_needed() return a bool
  ata: libata-transport: replace scnprintf with sysfs_emit for simple attributes
  ata: libata-eh: use bool for fastdrain in ata_eh_set_pending()
  ata: libata: Introduce ata_port_eh_scheduled()
  ata: libata-core: Rename ata_do_set_mode()
  ata: libata-eh: Rename and make ata_set_mode() static
  ata: libata-core: Make ata_dev_cleanup_cdl_resources() static
  ata: libata-core: Cache the general purpose log directory
  ata: libata_eh: Add debug messages to ata_eh_link_set_lpm()
  ata: libata-core: Reduce the number of messages signaling broken LPM
  ata: ahci: Disallow LPM policy control if not supported
  ata: ahci: Disallow LPM policy control for external ports
  ata: ahci: Disable DIPM if host lacks support
  ata: libata-sata: Disallow changing LPM state if not supported
  ata: libata-eh: Avoid unnecessary resets when revalidating devices
  ata: libata-core: Advertize device support for DIPM and HIPM features
  ...