| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Catch the case when a subdisk has no provider or no consumer | le | 2006-02-08 | 1 | -9/+12 |
| * | Get rid of the gv_bioq hack in most parts of the I/O path and | le | 2006-01-06 | 5 | -81/+89 |
| * | Don't pass error value pointer to g_read_data(9) at all if we don't | sobomax | 2005-11-30 | 1 | -3/+3 |
| * | Check for g_read_data(9) errors properly: | sobomax | 2005-11-30 | 1 | -1/+1 |
| * | Since we want a vinum geom created anytime the module loads, move | le | 2005-11-24 | 2 | -136/+27 |
| * | Whitespace. | le | 2005-11-20 | 1 | -8/+8 |
| * | Always declare variables at the start of the function. | le | 2005-11-20 | 1 | -47/+110 |
| * | Fix whitespace issue. | le | 2005-11-20 | 1 | -11/+11 |
| * | Finally bring in what was produced during Google SoC 2005: | le | 2005-11-19 | 6 | -4/+513 |
| * | Set the G_PF_WITHER flag on the subdisk provider that is about to | le | 2005-09-08 | 1 | -34/+1 |
| * | Prevent that sync operations can be started when they are already | le | 2005-08-28 | 1 | -71/+81 |
| * | Shuffle around the order in which the components are compiled. | le | 2005-08-26 | 1 | -14/+4 |
| * | Correct the check if a plex is accessible in case it is not up. | le | 2005-08-22 | 1 | -1/+1 |
| * | Make it possible to remove stale, left-over subdisks. | le | 2005-08-16 | 2 | -10/+11 |
| * | Fix a stupid logic bug introduced in geom_vinum_drive.c rev 1.18: | le | 2005-08-15 | 3 | -1/+9 |
| * | *) Implement round-robin reads for multiplex volumes. | le | 2005-07-15 | 2 | -25/+36 |
| * | When a drive dies, don't call g_wither_geom() directly, but instead | le | 2005-05-17 | 1 | -46/+71 |
| * | Only allow RAID5 plexes to be parity checked. | le | 2005-04-28 | 1 | -0/+4 |
| * | Remove test for zero sectorsize when tasting. This check doesn't | le | 2005-03-07 | 2 | -6/+0 |
| * | Don't allow to synchronize a plex that is already sychronizing. | le | 2005-03-04 | 1 | -13/+24 |
| * | Correctly calculate what to do and how to retry a request to a plex when | le | 2005-02-23 | 2 | -12/+47 |
| * | In case of drive errors, don't close the associated consumer and | le | 2005-02-17 | 1 | -10/+2 |
| * | Only report state changes of subdisks and plexes when there's | le | 2005-01-21 | 1 | -2/+14 |
| * | Don't initialize error with ENXIO as we might end up here when | le | 2005-01-21 | 1 | -2/+1 |
| * | Rename synchronization and initialization threads and prefix them | le | 2005-01-19 | 1 | -2/+2 |
| * | Although an object may already be known in the configuration, it's | le | 2005-01-19 | 3 | -5/+32 |
| * | Reset object flags after killing off an object's worker thread. | le | 2005-01-19 | 1 | -0/+6 |
| * | /* -> /*- for copyright notices, minor format tweaks as necessary | imp | 2005-01-06 | 5 | -5/+5 |
| * | Implement 'setstate' to allow setting the state of drives and subdisks | le | 2004-11-26 | 3 | -5/+82 |
| * | Implement checkparity/rebuildparity. | le | 2004-11-26 | 7 | -13/+316 |
| * | Move RAID5 offset calculation into a separate function to avoid | le | 2004-11-15 | 1 | -50/+56 |
| * | Share gv_roughlength() between kernel and userland, as we will need it | le | 2004-11-15 | 4 | -36/+36 |
| * | Give each plex a separate queue where held back bios are put on. | le | 2004-10-26 | 2 | -3/+16 |
| * | Don't allow to create a drive that already exists. | le | 2004-10-02 | 1 | -0/+8 |
| * | Correctly skip the '/dev/' part when creating new drives and prefix | le | 2004-10-02 | 2 | -5/+8 |
| * | Make it possible to rebuild degraded RAID5 plexes. Note that it is | le | 2004-09-30 | 6 | -16/+261 |
| * | Single concat or striped plexes don't need no special initialization | le | 2004-09-18 | 2 | -3/+21 |
| * | Re-vamp how I/O is handled in volumes and plexes. | le | 2004-09-18 | 9 | -663/+772 |
| * | Give the DRIVE geom a worker thread that picks up incoming bios, | le | 2004-09-13 | 5 | -49/+206 |
| * | Rename gv_kill_thread() to gv_kill_plex_thread(), since there are more | le | 2004-09-13 | 4 | -5/+5 |
| * | Save the config back to disk when a drive goes down. | le | 2004-09-13 | 2 | -1/+6 |
| * | Read a whole sector instead of GV_HDR_LEN, since a sector might be | le | 2004-09-13 | 2 | -2/+2 |
| * | Move config_new_drive() to the correct place and rename it to | le | 2004-08-27 | 3 | -32/+30 |
| * | When attaching a consumer from a volume to a plex, check if the | le | 2004-08-26 | 1 | -2/+21 |
| * | Compare the addresses of two RAID5 work packets directly instead | le | 2004-08-23 | 1 | -1/+1 |
| * | Implement the possibility to remove drives. | le | 2004-08-22 | 1 | -0/+90 |
| * | Add forgotten format specifier in a KASSERT and shut up the compiler. | le | 2004-08-22 | 1 | -1/+2 |
| * | A volume can be up if it has a degraded RAID5 plex. | le | 2004-08-19 | 1 | -1/+7 |
| * | Pretty print some informational messages. | le | 2004-08-18 | 1 | -1/+4 |
| * | Fix a stupid bug in the drive taste function: when checking if a | le | 2004-08-18 | 1 | -28/+10 |
