[dm-devel] SUSE multipath-tools patch resync
Hannes Reinecke
hare at suse.de
Wed May 18 15:03:28 UTC 2011
Hi Christophe,
finally I've finished rebasing my patch queue for
multipath-tools to the latest upstream git commit.
The resulting tree is at:
git.kernel.org:/pub/scm/linux/kernel/git/hare/multipath-tools.git
branch for-christophe
Here are the summary lines of the patches:
* Reload map for device read-only setting changes
* Increase priority value for emc priority callout
* Allow dev_loss to be set to 'infinity'
* Update manpages
* multipath: add '-t' option to dump internal hwtable
* Reassign existing device-mapper maps
* alua: Handle LBA_DEPENDENT state
* Update multipathd init script for SuSE
* multipathd: Add PID to 'show daemon' cli command
* Update pid file handling
* Serialize startup on large machines
* Add 'shutdown' cli command
* multipathd: Update 'max_fds' handling
* multipathd: crash in reconfigure CLI command
* Add 'max_polling_interval' config variable
* libmultipath: Only count UP and GHOST paths for prio update
* Always synchronize with dm state
* multipathd is not starting waitevent checker for single paths
* More debugging output when synchronizing path states
* libmultipath: Make path state message unique
* Update dev_loss_tmo for no_path_retry
* Check for valid argument in update_multipath_strings()
* libmultipath: Remove duplicate calls to path_offline()
* Check for offline path in get_prio()
* Use state name in get_state()
* Only check offline status for SCSI devices
* libmultipath: zero out sense buffer in do_inq()
* multipathd: fix memory issues in cli.c
* multipathd: Remove handling of 'umount' events
* Safe memory allocation in cli_handlers
* multipathd: Fix uxlsnr race condition on shutdown
* Rework sysfs device handling in multipathd
* multipath: add SuSE init file
* Checker name is not displayed on failure
* Unload priority modules
* Display avg priority as group priority
* libmultipath: Improve debugging of log messages
* libmultipath: rework sysfs handling
* libmultipath: update waiter handling
* libmultipath: do not fallback to search /proc/partitions
* Start uevent service handler from main thread
* Use lists for uevent processing
* Remove sysfs_attr cache
* Fixup strip trailing whitespaces for getuid return value
* Check return value for select_alias()
* Use enum free_path_mode for free_paths argument
* vector paranoia
* Update cli request processing
* Implement 'bindings_file' option
* kpartx: Intendation fix in devmapper.c
* Use regmatch when checking for duplicates in hwtable
* Update hwtable factorization
* Specify prioritizer in the multipath section, too
* Static Load balancing prioritizer
* Synchronize dm states for IO errors
* Add 'max_polling_interval' config variable
* libmultipath: do not access dm structures after dm_task_destroy
* Correctly remove logical partition maps
* rdac checker: Whitespace cleanup
* libmultipath: Fix possible string overflow
* More debugging output for feature changes
* Fixup debugging in dmparser.c
* Correctly update feature strings
* Select 'features' from multipath configuration
* Increase uevent buffer size
* libmultipath: remove strcmp_chomp()
* Remove sysfs_lookup_devpath_by_subsys_id()
* libmultipath: resolve hash collisions in pgcmp()
* libmultipath: correct path count in setup_map()
* Make status variable local
* Make params variable local
* multipath segfaults if configuration file contains errors
* Use '--replace-whitespace' option for scsi_id
* Asynchronous mode for tur checker
* Move setup_thread_attr() to uevent.c
* TUR checker should not return 'failed' for reservation conflict
* directio: Reset 'running' parameter
* Use timeout parameter for directio
* libmultipath: check argument length in execute_program()
* readsector0: block count calculation wrong
* checkers: argument paranoia
* Use refcounting for checkers
* kpartx: Add option '-f' to force devmap creation
* Add 'no_partitions' feature
* Implement 'update' option for kpartx
* kpartx: Indentation fix
* Add compability flag LIBDM_API_COOKIE
(Did I mention I had quite a few outstanding?)
They are ordered in reverse order, in case you'd want to
cross-check against the git tree.
How to proceed best here?
Sending them all for review to the mailinglist seems a
bit excessive, but then that'll be automated anyway,
so I don't care.
Just tell me what you prefer.
Cheers,
Hannes
--
Dr. Hannes Reinecke zSeries & Storage
hare at suse.de +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: J. Hawn, J. Guild, F. Imendörffer, HRB 16746 (AG Nürnberg)
More information about the dm-devel
mailing list