summaryrefslogtreecommitdiff
path: root/sys/dev/raidframe
AgeCommit message (Collapse)Author
2002-03-08Pool deals fairly well with physical memory shortage, but it doesn'tthorpej
deal with shortages of the VM maps where the backing pages are mapped (usually kmem_map). Try to deal with this: * Group all information about the backend allocator for a pool in a separate structure. The pool references this structure, rather than the individual fields. * Change the pool_init() API accordingly, and adjust all callers. * Link all pools using the same backend allocator on a list. * The backend allocator is responsible for waiting for physical memory to become available, but will still fail if it cannot callocate KVA space for the pages. If this happens, carefully drain all pools using the same backend allocator, so that some KVA space can be freed. * Change pool_reclaim() to indicate if it actually succeeded in freeing some pages, and use that information to make draining easier and more efficient. * Get rid of PR_URGENT. There was only one use of it, and it could be dealt with by the caller. From art@openbsd.org.
2002-03-04Correct misspellings of "failed".wiz
2002-01-19Bah.. Not sure how those ";"'s got missed :-(oster
2002-01-19Let's not bother building a number of infrequently used RAID types.oster
The code is still there if people want it, but these other RAID types (using various DECLUSTERING and SPARING techniques) arn't being used in practise. Saves about 100K on i386 GENERIC kernels.
2002-01-19Only include some debugging printf's if DEBUG is defined.oster
2002-01-09We only need a single component buffer pool.thorpej
2002-01-09Move a bunch of debugging stuff to be only used if DEBUG is turned on.oster
2002-01-07Revert previous. Need to be more careful with what gets chopped out.oster
2002-01-07Some of this stuff is really just for debugging, so compile it only ifoster
DEBUG is defined.
2002-01-07#if 0 rf_SetReconfiguredMode() since it is unused now, but may beoster
needed someday.
2001-12-16Fix an off-by-one error when autoconfig is hunting for a free RAIDoster
device in the rare event that the one it wants is already in use. Thanks to Wolfgang Stukenbrock for noticing the bug and filing the PR. This fix addresses PR#14862.
2001-11-28use #ifdef RAID_AUTOCONFIG (for consistency with the rest of the kernel)lukem
2001-11-20Cleanup some whitespace that indent got wrong a long time ago.oster
2001-11-15don't need <sys/types.h> when including <sys/param.h>lukem
2001-11-13add RCSIDslukem
2001-10-06Backout a line that accidentally got included when the .h reorg went in.oster
2001-10-05More #if 0's bite the dust.oster
2001-10-05Nuke some #if 0 stuff.oster
2001-10-04Tidy up the #includes a bit. (coulda/shoulda done that with last commit.)oster
2001-10-04rf_options.h isn't needed here.oster
2001-10-04If rf_options.h needs to be include it, then include it. Don't rely onoster
another .h file to pull it in.
2001-10-04Nuke a couple of unused extern variable declarations.oster
2001-10-04These have been completely replaced by raidframevar.h and raidframeio.h.oster
2001-10-04Setup installation mechanism for <dev/raidframe/raidframevar.h> andoster
<dev/raidframe/raidframeio.h>
2001-10-04Step 2 of the disentanglement. We now look to <dev/raidframe/*> foroster
the stuff that used to live in rf_types.h, rf_raidframe.h, rf_layout.h, rf_netbsd.h, rf_raid.h, rf_decluster,h, and a few other places. Believe it or not, when this is all done, things will be cleaner. No functional changes to RAIDframe.
2001-10-04Step 1 of the disentangling of RAIDframe's .h files:oster
raidframevar.h - contains bits of rf_types.h, rf_layout.h, rf_netbsd.h, rf_raid.h, and rf_decluster.h. raidframeio.h - contains the bits needed for doing IOCTL's w/ RAIDframe. These bits will be visible to userland.
2001-09-26RF_DEV2RAIDID isn't used anywhere anymore. Bye bye.oster
2001-09-26- rf_types.h doesn't really need to pull in rf_archs.hoster
- for *.c, don't rely on rf_types.h to pull in rf_archs.h
2001-09-25Nuke an unused typedef.oster
2001-09-16Spell 'occurred' with two 'r's.wiz
2001-09-01Disable all of the RF_ASSERT()s by default, enabling them if thethorpej
RAID_DIAGNOSTIC option is specified in the kernel configuration file.
2001-07-27Nuke the 5 NULL parameters from rf_DoAccess(). Observe that 4 ofoster
the variables are passed directly to rf_AllocRaidAccDesc(), and the 5th isn't even used. Hunt them down and nuke them too.
2001-07-18bcmp -> memcmpthorpej
2001-07-18bzero -> memsetthorpej
2001-07-16If we're closing the last partition on a RAID set and the system isoster
being shutdown, then unconfigure the RAID set too. This fixes a number of issues with doing proper unconfigures/shutdowns of multi-level RAID sets. Thanks to Jason Thorpe and Bill Squier for the ideas/suggestions on how/where to do this, and to Bill Squier for testing.
2001-06-21We don't need to check for raidautoconfig here, so nuke it andoster
clean up a little.
2001-06-20#if out a few more bits that arn't required if we are not doingoster
PQ or RAID6.
2001-06-19A bunch of these bits are only needed for the EVENODD stuff.oster
Shaves a little better than 1K off the size of an i386 kernel.
2001-06-14It's silly to need a parity rebuild after a reconstruction has completed.oster
If we've just reconstructed a disk, then the parity is known to be correct. (XXX doesn't hold for RAID 6!)
2001-06-08#endif /* takes no argument */mrg
2001-04-05A value of 1 for tracks/cylinder in the disklabel is apparently sub-optimal.oster
Thus we choose "4 * number_of_columns" as a more reasonable value (until someone comes up with something better). This pretends to properly address PR#11989.
2001-02-03include <sys/param.h> before <sys/ioctl.h>mrg
2001-01-27Un-'__P'ify.oster
2001-01-27Nuke some bogus debug code that was already '#if 0'ed out.oster
2001-01-27Cosmetic cleanup of some comments, and nuke an unused #define.oster
2001-01-27Having moved some of the bits needed by raidctl into rf_configure.c, weoster
no longer need those bits here. In particular, this file is now only used in the kernel, and that allows us to clean up a lot of cruft.
2001-01-27More #if's. This needs some serious rototillage, but a few #if's willoster
suffice for now.
2001-01-26My.. what a twisty little maze of #defines and dependencies. Some of thisoster
code is used for more than anticipated, and I missed a few checks last night. Thanks to Bernd Ernesti for poking me about this.
2001-01-26Nobody calls these bits if we arn't doing RAID5 with rotated spare stuff.oster
Add an #if.
2001-01-26These bits are only needed for the interdecluster stuff.. wrap themoster
with an #if.