summaryrefslogtreecommitdiffstats
path: root/sys/geom
Commit message (Collapse)AuthorAgeFilesLines
* Avoid double-free panic.phk2003-05-051-1/+0
| | | | Tripped up: DougB
* Re-order the the initialization slightly to improve structure.phk2003-05-051-22/+22
|
* Use a dedicated malloc(9) bucket for sector storage.phk2003-05-051-7/+9
|
* Don't warn if the rawoffset is zero, that is actually the best value itphk2003-05-051-1/+1
| | | | could have.
* Turn the check that rawoffset == mbroffset into a warning instead.phk2003-05-051-3/+5
|
* Only accept a rawoffset if it is identical to the mbroffset.phk2003-05-041-0/+4
|
* Add a way to read the current mbroffset from a BSD label class.phk2003-05-041-1/+7
|
* Add gctl_set_param() function.phk2003-05-042-0/+26
|
* Remove debugging printfs which should not have been committed.phk2003-05-042-7/+0
|
* Add a OAM interface for changing the label and writing the boot code.phk2003-05-031-26/+126
|
* remove unused variables.phk2003-05-031-2/+0
| | | | Spotted by: dougb
* Make bsd_disklabel_le_enc calculate the checksum and fill it in.phk2003-05-022-268/+149
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (If there is a legitimate need to correctly encode and pack a disklabel with an invalid checksum custom tools can be built for that.) Make bsd_disklabel_le_dec() validate the magics, number of partitions (against a new parameter) and the checksum. Vastly simplify the logic of the GEOM::BSD class implementation: Let g_bsd_modify() always take a byte-stream label. This simplifies all users, except the ioctl's which now have to convert to a byte-stream first. Their loss. g_bsd_modify() is called with topology held now, and it returns with it held. Always update the md5sum in g_bsd_modify(), otherwise the check is no use after the first modification of the label. Make the MD5 over the bytestream version of the label. Move the rawoffset hack to g_bsd_modify() and remove all the inram/ondisk conversions. Don't configure hotspots in g_bsd_modify(), do it in taste instead, we do not support moving the label to a different location on the fly anyway. This passes all current regression tests.
* Pull in bcopy() prototype from <string.h> when compiled in userland.phk2003-05-021-0/+2
|
* Considering that I did cast the arguments to (intmax_t) I must havephk2003-05-021-4/+4
| | | | been sleepy since I used %qd instead of %jd.
* Style improvement.phk2003-05-021-4/+3
|
* Use g_wither_geom() and plug memory leaks.phk2003-05-021-4/+4
|
* Plug memory leaks.phk2003-05-021-0/+16
|
* Use an uma-zone for allocation bio requests.phk2003-05-021-13/+10
|
* Use g_slice_spoiled() instead of g_std_spoiled().phk2003-05-021-1/+2
| | | | Add XXX comment about minor memory leak until I can fix it.
* Use g_slice_spoiled() instead of g_std_spoiled().phk2003-05-021-1/+1
|
* Use g_slice_spoiled().phk2003-05-021-3/+5
| | | | Free buffer from g_read_data().
* Back out all the stuff that didn't belong in the last commit.phk2003-05-028-45/+27
|
* Use g_slice_spoiled() rather than g_std_spoiled().phk2003-05-029-28/+47
| | | | Remember to free the buffer we got from g_read_data().
* Use g_slice_spoiled() not g_std_spoiled()phk2003-05-021-1/+1
|
* Use g_slice_spoiled() rather than g_std_spoiled()phk2003-05-021-2/+2
|
* Use g_slice_spoiled() rather than g_std_spoiled().phk2003-05-021-1/+1
|
* Use a more tailored spoil routine for slices, and take advantage ofphk2003-05-022-15/+20
| | | | g_wither_geom() to do most of the work for us.
* Style improvement.phk2003-05-021-4/+2
|
* Use g_wither_geom() for cleanup.phk2003-05-022-6/+4
|
* Rework the "withering" mechanism:phk2003-05-022-17/+47
| | | | Introduce g_wither_geom() to do the work in one single place.
* Rename g_slice_init() to the more appropriate g_slice_alloc() and givephk2003-05-021-4/+14
| | | | it a g_slice_free() partner function.
* style improvement.phk2003-05-021-3/+2
|
* Get rid of trivial function g_destroy_event().phk2003-05-021-10/+3
|
* Plug some memory-leaks.phk2003-05-011-2/+7
|
* Remove the now obsolete geomidorname hack.phk2003-05-012-107/+0
|
* Add a new flag, EV_CANCELED, and use it to make g_waitfor_event() returnphk2003-05-011-1/+5
| | | | EAGAIN if an event got canceled.
* When events on a reference is cancelled, check our doorstep first,phk2003-05-011-0/+7
| | | | it might be an orphan.
* Remove now unneeded special case for "geom.ctl".phk2003-05-011-2/+0
|
* Remove DIOCGPC98 ioctl.nyan2003-05-011-5/+0
|
* - Move decoding pc98_partition function into geom_pc98_enc.c.nyan2003-05-012-24/+79
| | | | - Add encoding pc98_partition function.
* Don't emulate a MBR by handling the MBR::type attribute. It ismarcel2003-05-011-21/+1
| | | | | not needed at all. The BSD class will attach to a GPT class without it.
* Fix some easy, global, lint warnings. In most cases, this meansmarkm2003-04-301-1/+1
| | | | | making some local variables static. In a couple of cases, this means removing an unused variable.
* Fix an obscure fencepost error in GBDE's sector mapping code:phk2003-04-291-0/+26
| | | | | | | | | | | | | | | For certain combinations of sectorsize, mediasize and random numbers (used to define the mapping), a multisector read or write would ignore some subset of the sectors past the first sector in the request because those sectors would be mapped past the end of the parent device, and normal "end of media" truncation would zap that part of the request. Rev 1.19+1.20 of g_bde_work.c added the check which should have alerted me to this happening. This commit maps the request correctly and adds KASSERTS to make sure things stay inside the parent device. This does not change the on-disk layout of GBDE, there is no need to backup/restore.
* Typo in last commit: Do not press xZZ to leave vi(1).phk2003-04-291-1/+1
|
* When a bio comes back from below with a zero error code, check thatphk2003-04-291-0/+4
| | | | | | | it wrote the full length. The only case where this should be able to happen is if we try to read/write past the end and the request is truncated. We obviously should never try to do that, so this code should never activate.
* Deprecate machine/limits.h in favor of new sys/limits.h.kan2003-04-291-1/+1
| | | | | | | Change all in-tree consumers to include <sys/limits.h> Discussed on: standards@ Partially submitted by: Craig Rodrigues <rodrigc@attbi.com>
* I accidentally leaked this debugging tool in with my last commit.phk2003-04-281-2/+4
| | | | Disable it with a direct warning.
* Rename g_bde_get_sector() to g_bde_get_keysector() and pick up thephk2003-04-282-5/+42
| | | | offset from the work packet.
* Only attempt total cache-purge once in case of failure.phk2003-04-281-4/+3
|
* Better criteria for skipping disk reading BIO_READ work packets.phk2003-04-281-1/+2
|
OpenPOWER on IntegriCloud