summaryrefslogtreecommitdiff
path: root/sys/dev/raidframe
AgeCommit message (Expand)Author
2001-01-26Whoops.. Leave them on.. (for now.)oster
2001-01-26Give the parity declustering code their own 'RAID arch' types, andoster
2001-01-26Lots of stuff here that is only used for some of the 'fancier' RAID types.oster
2001-01-26Clean up some INDENT warnings. (yes, these have been there for a while!)oster
2001-01-26Various parts of this file are only needed by the even/odd and PQoster
2001-01-26Don't build the stuff in this file if RF_INCLUDE_EVENODD is not defined.oster
2001-01-26Ensure we update the 'partitionSize' field of the component labelsoster
2001-01-08Return error in the case of using ODIOCGDINFO or ODIOCGDEFLABEL whenfvdl
2001-01-07Adapt all disk devices in MI directories to handle ODIOC* callsfvdl
2000-12-15For the RF_ASSERT() and RF_PANIC() macros, call a function to do theoster
2000-12-05Fix a couple of warnings about uninitialized variables. Thanks go tooster
2000-12-04Comment out 2 cases of unitialized variable use (only used whenfvdl
2000-11-20rationalize the use of b_flags for geteblk() buffers.chs
2000-10-20Move disk_busy() and disk_unbusy() to more sane locations. Valuesoster
2000-10-17Don't touch a failed component at all. Thanks to Robert Elz (kre) foroster
2000-10-12Minor fixup for a printf(). Noted by Robert Elz.oster
2000-09-21convert tsleep()->ltsleep(), which makes the kernel complain lessoster
2000-09-21Correct a few calls to VOP_CLOSE().oster
2000-09-11Be consistent: use mcpair->cond instead of mcpair->flag.oster
2000-09-11Only reset the timer value if it actually accumlated time.oster
2000-09-08Make sure we are passing a locked vp to VOP_CLOSE.oster
2000-08-20Use RF_WAIT_COND() to avoid locking errors.thorpej
2000-08-20RF_COND_WAIT(): use ltsleep().thorpej
2000-08-19Remove a bogus VOP_UNLOCK.oster
2000-08-19Don't try to initialize a component label on a failed component.oster
2000-07-14Fix the status reporting for copybacks. Thanks to Hubert Feyrer foroster
2000-06-29Nuke a printf that *I* added at least as early as Nov, 1998, and which isoster
2000-06-11Nuke a (mostly) now-incorrect comment.oster
2000-06-04Remove outdated comment.oster
2000-06-04Merge rf_update_component_labels() and rf_final_update_component_labels().oster
2000-06-04Minor prototype cleanup.oster
2000-06-03Allow getting a component label from a spare disk.oster
2000-06-02Use 'componentN' rather than leaving blanks for non-existant disksoster
2000-06-01Make booted_device global (and booted_parition for consitency).matt
2000-05-31Oops.. reconstruction percentages were being reported incorrectly.oster
2000-05-30For a RAID 1 set that has N components, allow the set to auto-configure asoster
2000-05-30Use #include <sys/kernel.h> instead of a 'extern int hz'.oster
2000-05-29Don't let the autodetection code get tripped up by negative mod-counters.oster
2000-05-29Fix a couple of typos. (Thanks Wiz! :) )oster
2000-05-28In the event that an up-to-date component cannot be located for a specificoster
2000-05-28We need to be slightly more clever during auto-configuration, to makeoster
2000-05-28Umm.. Complete is not equal to 'left to do'. Fix the math.oster
2000-05-28Abort any parity re-writes before unconfiguring a RAID set.oster
2000-05-28When setting auto-config/rootable bits, only try to update theoster
2000-05-28- Add a mechanism for obtaining finer-grained 'progress' informationoster
2000-05-27Components which match on everything but the mod_counter areoster
2000-05-27Add a comment to clarify why we actually want the component label tooster
2000-05-27Checking parity re-write status makes no sense on RAID 0, but let'soster
2000-05-23Const poison the layout name.thorpej
2000-05-19A foolish consistency; most parts of the kernel use bp->b_data, sothorpej